27 Burst results for "Kathleen Mulch"

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:29 min | 3 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"Your host, Kathleen mulch. And I'm your host Ronald smells her. And on today's podcast, we're going to actually continue along with something we had done in a previous podcast. We're talking about some of the hot and growing markets and the data and AI, machine learning, cognitive tech automation, analyst analysis, sort of parts of the analytics. I mean, part of the markets. And if you are new to the AI today podcast, you might not know that we've been spending a lot of time over this past over 240 or so episodes talking about where these AI markets are going. A lot of it based on the research that we are producing here at cognitive as part of our cog axis, which is our market intelligence that we do. We track well over 20,000 vendors in this crazy AI marketplace. Expanding now to the data automation and analytics marketplace. But we had over 20,000 AI vendors before we expanded even to automation and analytics and as we observe these markets and we track for our customers, which are primarily large enterprises and government agencies as well as technology vendors and solution providers. We see changes in these markets as companies are putting AI into practice. And as we see these changes, well, we're sharing some of those insights with you here on our AI today podcast. Exactly. We had talked about how in 2022, we wanted to focus our podcasts more on education, as well as highlight some of our research that we do in the space. You know, as Ron mentioned, we're analysts, at cognitive, which is now focused beyond just artificial intelligence and machine learning to also data automation and analytics. And we said, you know, we should share with our podcast listeners, some of you are cog access subscribers, which is to our market intelligence. Subscription. But some of you aren't. And so we wanted to make sure that you still can gain insights from this podcast. So in a previous podcast, we had talked about the data labeling space. That is a hot market that we have been covering since 2019. We are the only analyst firm on the market who is doing pretty much any research in this space, but definitely to the extent and the depth that we go. So we encourage you to listen to that podcast if you haven't already. And I will make sure to link to it in the show notes. When we had produced our snapshot, which was in December of 2021, originally, we had synthetic data as part of the overall data labeling landscape. And then as we continued to do more research into this space and look at the vendors there, we said, you know what? This really deserves to be its own market and break it out into its own snapshot report and coverage area. And so that's what we have since done. We also had a podcast recently that talked about our data infrastructure, the general area of that. And in that podcast, we also talked about how we broke out synthetic data. But we wanted to devote today's podcast to synthetic data to talk to our audience about what it is. The importance of it and why we're covering it. Yeah..

Kathleen mulch Ronald Ron
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:33 min | 4 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"Host, Kathleen mulch. And I'm your host Ronald Schmidt. And on today's podcast, we're going to dive deep again into our education series. This was actually very popular. We noted one from your comments. So thank you very much to our AI today podcast listeners. We really do encourage you to reach out to us. We've been doing the AI today podcast now for. This is going in our 5th year, 240 plus episodes. And it's always important to hear from our listeners as to what they really want to hear from. You could be listening to any sort of podcast out there and certainly there's lots of AI podcasts out there. And what we found is that as you know, one of the things that we do a cognate is we provide education and training. And as we were sharing some of that with our listeners, you all responded very positively by letting us know some of you actually signed up for our CPU Mayo training, which is great. That's not really necessarily the intention of doing this. We just wanted to share with our listeners that education. And of course, we saw our listener numbers go up. So that was the big sign that you're interested. So we're going to keep doing education stuff. Exactly. And in 2022, we really want to focus on continuing to provide valuable podcasts valuable content. And like Ron said in 2021, we did our AI education series, and that was incredibly popular. So this podcast is going to be a continuation of that. But we also did an AI failure series. And if you haven't checked that out, I definitely encourage you to listen to those. That was talking about common reasons we see AI projects fail and how to avoid that so that you don't make the same mistakes. If you're not already subscribed to the AI today podcast, I encourage you to subscribe so that you're notified of all of our upcoming podcasts. We publish at least once a week every Wednesday. And sometimes we have extra bonus episodes during the week as well. So definitely stay subscribed so that you can catch up on all of that..

Kathleen mulch Ronald Schmidt Ron
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

01:56 min | 4 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"I'm your host, Kathleen mulch. And I'm your host Ronald Schmidt. And welcome to 2022 for those of you that are listening. Hopefully you've heard our forecasts on the market that we did in general on the AI and the AI space and waste ways things are going. You know one of the things I noticed on that one if you listen to it is we didn't really spend too much time on where we think artificial general intelligence is going. And maybe we will do an update on that on some of the more kind of forward looking things that we don't necessarily cover in our research. And for those of you that are not aware, Kathleen and myself, we're managing partners of an analyst firm called cognate, which focuses on research advisory and education on four major markets. The markets for cognitive technologies, which include AI, we obviously spend a lot of time on that. On automation, which helps support many of the goals for intelligence, but of course, is not automation is not intelligence. And then the third area is advanced analytics. We're trying to gain more information and knowledge and value from our data over time. So we talk about that. The fourth area is around digitization, which is taking the things that are not digital and making them digital. And we produce market intelligence and market research. And what we're going to do with you, we're going to do much more of this in 2022 and beyond is we're going to share from you with you. Some of the things that we are seeing in these various markets as customers like yourselves on this podcast and vendors maybe some of you listening as well, as you're trying to get in and use these things, how are these markets changing? So you can make use of this great technology. Exactly. One thing that we said we really wanted to focus on with this podcast for 2022 was education. Because a lot of feedback that we got from the podcasts that were listened to, you know, when we looked at data and also from our listeners as well, who reach out to us, so thank you for reaching out. And everybody is always welcome to reach out to us. Info at cognitive dot com..

Kathleen mulch Ronald Schmidt Kathleen
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:40 min | 4 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"The AI today podcast produced by cognac cuts through the hype and noise to identify what is really happening now in the world of artificial intelligence. Learn about emerging AI trends, technologies, and use cases from cognitive analysts and guest experts. Hello, and welcome to the AI today podcast. I'm your host Kathleen mulch. And I'm your host Ronald Schmidt. And welcome for those of you that are listening to our podcast when we just come out with and welcome to 2022. For those of you that are listening to this later in the year, you're probably thinking it's been 2022 already. Why am I even listening to the welcome? But I had to say for those real estate podcasts, one of the things that we like to do at the beginning of every year, is really provide some insights and forecasts and trends as to where we think things are going to happen in this upcoming year. I think one of the things that we're going to do a little bit more differently I think on this podcast and we may have in previous ones is instead of trying to come up with some crazy out of the blue predictions on whatever we're going to actually provide some insights into our market intelligence and what it is actually saying about the many different corners of the AI automation data and analytics markets that we are tracking and give you an idea from that from what our research is telling us, where we think things are heading into this upcoming year. Exactly. We always like these podcasts because it gives us a time to look back and reflect on what we've seen. And then also make some forecasts and trends and kind of look to say, okay, where have the markets been and where do we think that they are going in 2022? So if you're not familiar with us by now, if this is one of your first times joining the podcast, the AI today podcast has been going strong for many years now. We're well into season 5. And we are running myself our analysts that cognitive, which is an AI focused research advisory and education firm..

Kathleen mulch Ronald Schmidt
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:24 min | 5 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"The AI today podcast produced by cognitive cuts through the hype and noise to identify what is really happening now in the world of artificial intelligence. Learn about emerging AI trends, technologies, and use cases from cognitive analysts and guest experts. Hello, and welcome to the AI today podcast. I'm your host Kathleen mulch. And I'm your host Ronald Schmidt. And this is the tenth of ten podcasts in our AI failure series. So if you've been listening to our AI today podcast over the past several months, actually, that we've been doing this. We've been talking about some of the things you might be hearing in the news and perhaps from others in the industry that a lot of AI projects are failing to have a very high failure rate according to the common press. When people are saying. And that's certainly not false. I mean, that is true in that in many cases. These AI projects that people set about to start, they're trying to solve any of a number of problems. We go into that, by the way, in other podcast episodes we talk about all the different kinds of problems people are trying to solve with AI. And if this is your first podcast, if this is the very first AIA podcast, you're hearing, you should know, hey, first of all, there are 9 other episodes in this particular series. But in general, we have over 230 some odd episodes that we've been doing for the last four years. This is our 5th season now going into 5 years of doing AI today and the typical thing that we do on AI today if you're not listening to the failure series is we usually talk about AI successes. So we're not pessimistic or anything like that. We do believe and we have actually seen AIB successful. In many cases and we have interviewed many people who have been successful with AI. And in large businesses and government agencies, an organization across the world from in technology companies. And everybody..

Kathleen mulch Ronald Schmidt AIB
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:06 min | 5 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"To the AI today podcast. I'm your host, Kathleen mulch. And I'm your host Donald Trump, and thank you for joining us on yet another episode of AI today we're well over 200 episodes now into our 5th season. Maybe this is actually officially our 5th season and we started in September of 2017. So whatever gifts like the end of the government fiscal year, you're heading into fall, the kids are in school. That's what I know. I think it's time for a new season. So you're listening to the AI today podcast for the first time, perhaps you're coming from one of our other podcasts that we've done are some of our partner podcasts. You should know that this is our AI today. We've been going strong for four years interviewing folks on what is actually happening with AI today, really focusing from your perspectives as those of you many of you who are listening are trying to put AI to practice. We love the promise of AI and the thought of intelligent machines and all the things that we can do, stuff that we talk about in our podcast with so called so called 7 patterns of AI. Conversational systems and recognition systems, patterns and anomalies and predictive analytics, hyper personalization, goal driven systems, and, of course, autonomous systems, you're trying to put them into practice. It's nice to know what other people are doing as well. So if you haven't yet had a chance, please do, subscribe to this podcast and listen to our many interviews that we've had with large enterprise is small enterprises, international governments, local governments, and organizations in between. So we are really excited to have with us today. Gilman Louie, who is the cofounder and partner of ulsa, Louis partners, and early stage technology venture capital firm and commissioner for the national security commission on AI our second podcast on the subject. So welcome so much gillen for joining us here on the AI today podcast. I'm so excited to be here. Thank you for inviting me. Yeah, thanks so much for joining us today. We're very excited to have you and looking forward to this interview very much, especially because you have a very diverse background. So we'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background more than we gave in that quick introduction. And how you got involved with.

Kathleen mulch Donald Trump Gilman Louie ulsa Louis partners national security commission o gillen
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:45 min | 7 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"Welcome to the ai today. Podcast i'm your host kathleen mulch and i'm your host wallich knows or and this podcast episode continues what we did a podcast episode or two ago where we were talking about the reasons why. Ai projects fail. And if you're just hearing this podcast for the first time. Amy you just tuned into today for the first time which gets welcome. You are part of a great family. Podcast we really try to focus on the here and now making a i work you know. Our perspective is on the people who are dealing with. Ai today we we love the future of what can be. We love the promise. We love the the all the great things you hear about the research and science fiction and what the vendors are saying but on this podcast we focus on the here and now today people actually trying to make the stuff work and well. You may not be surprised but people are having a hard time. There is a high failure rates of ai projects. And well we're going to get into. Why exactly those products are feeling exactly. So you know. If you're not familiar with us. Ron and i are with cognreznick run. Ai focused research advisory in education firm. So we have been around for many years and so has our podcast so throughout the series Throughout the course of pag melinda we've talked to many many vendors we've talked to many end users and enterprises government organizations on how they're actually you know eh guy today so how they're actually implementing a today and over the course of these many years. We have seen a number of reasons why projects fail and you know in general. There's some general themes and so that we thought it was important to have this. Ai failure series with the podcast. Because we wanted to talk about you know. What are the reasons that we're seeing letter listeners. Know these reasons and then why they're occurring and how you can help to overcome them so in today's episode we really wanna talk about One of the reasons we've seen my i..

kathleen mulch wallich pag melinda Amy Ron
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

02:19 min | 9 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"Can you explain how you picked the name for your podcast. I think it goes back to those experiences. David just reminded me about simply go to a makeup group and explain how we were doing it. People were interested in how to get started with data science in their particular context. But also we were going to be ups speakers but as Audience members. And i remember very well one at a lodge technology company running a social network and i tried. I tried to ask one questions to one of the scientists. I asked about the difficulties of getting hold of the right data for your question and this polite intimate Listen to my question. They looked at me. Like i was from a different planet. And then he said well. We have a team of data. Engineer is dedicated to that. And if there's something that we need we don't have we just pig message on slack. Command like achievement. Few later they send us link to the st bucket where that they tours and i just realized how that said we have some of the same woods and add job title. Maybe we have some of the same training and maybe some of what we do. Some of it will look similar day to day but really the context you're operating is is quite different and when i interviewed for that job i think we got talking about this issue. I recently published something. Actually computer weekly On the psycho. Don't just listen to the one percent view of data science and david you remembered An article that you had read about was code. Doc netted develops what was the. Yeah that that that resonated with us it was about a software developers who aren't like working kind of hipster startups. They were just working on old enterprise software and just making it tick along and essentially powering the entire world. But you won't find them at the cool meet ups they just go to work work with some really old version of and then go home

kathleen mulch ronald schmeltzer acog melissa sean mcgurk David
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

02:01 min | 10 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"We'd like to start by having you introduce yourself to our listeners. Tell them a little bit about your background and also why you started your podcast. Sure yeah so. I am under a crank kavas insured. I'm currently ferragu. Almost four fewer phd student at stanford Focusing on a work a lot on robotics reinforcement learning and Yeah Had kind of an interesting experience. Where early on my phd A lot of this hype a lot of ridiculous new stories around i and so are on. The time has started the sink called skynet today which is kind of a funny title but basically we had explain articles in tried to defuse high. So yeah that has become an ongoing project and last year You're thinking well. How can be really expand and we check raider audience and luckily sharon was also at stanford a good friend and also interested in doing something related to this end way to podcasting so he started stock ai to interview some people also to discuss a lot of news. So sharon also go ahead. Let us know who you are. Yeah feels like eons ago when we started it because it was pre pandemic But actually right before the pandemic And it was kind of a cool way to stay up to date with the news but also a trying to demystify it since there is so much hype going around. Not just you know in the research world but also in media and mainstream media My background is. I just received my doctor and a i Advised by andrew ing on. Who is a professor and also An ai i guess. Thought leader in the space And i think a huge reason why we started the podcast was to you know dispelled all that hype on given our knowledge and our background of what actually works in a.

cog melinda cotonou kathleen mulch ronald schmeltzer
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:11 min | 10 months ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"The today podcast produced by cog melinda cuts through the hype and noise to identify. What is really happening now. In the world of artificial intelligence learn about emerging trends technologies and use cases from cotonou -lica- analysts and guests experts. Hello and welcome to the ai today. Podcast i'm your host kathleen mulch and i'm your host ronald schmeltzer and we have a treat for you today..

cog melinda cotonou kathleen mulch ronald schmeltzer
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

01:46 min | 1 year ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"So you know. We focus heavily on art. Official intelligence machine learning cognitive technologies in our education but one area where we're spending some time. Educating our customers is about robotic process automation are and how it fits into the overall. Ai picture we thought it was important to do this. Because you know especially a few years ago. I think there was some confusion in the market and maybe You know the confusion wasn't always meant to be corrected by some of the vendors about our pa you know how they're similar how they're not similar and how they fit with each other so are are not intelligent and hopefully you know we have hammered that in with a bunch of podcasts that we've had and in our education do as well and they're not meant to be intelligent but they certainly do at a lot of You know value to what you're doing and you can add intelligence to them to make them even more useful and we go over different levels of how you can add to your rpi bots. The real value that are pa provides it comes from removing the bought from the human. You know no longer are humans. You take away those tasks that huge just aren't good at you know. We are not good at sitting and doing very repetitive tasks over and over and over for hours every day so rpi helps take that out of the human take that repetitiveness out of the human and then we give them to machine so that they can do it. These repetitive tasks take time. They can be very boring. They also are error-prone as well. Because as i mentioned humans are not meant to be doing that all the time

kathleen mulch ron schmeltzer one few years ago today one area things cowed Cognreznick dot com few months dot
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

02:32 min | 1 year ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"So we're so excited to have with us today. Joy bond guerrero. Who's the chief data officer of california. Hi joy and thanks so much for joining us. Thank you for having me excited to be here today. We'd like to start by having you introduce yourself to our listeners. Tell them a little bit about your background. And your current role is chief data officer for the state of california. Sure so. I always like to describe myself as a four legged stool also known as the chair ends so my background actually got started in data design and technology so the first three legs of the stool and not was on designing and developing data and information systems in the city in parish of new orleans. Both pre and post katrina and what i learned along the way. There was that the way. We're developing those systems if we grounded them in user needs made the music music center. They could have high impact on in They could reshape public policy choices. And so really i. I sort of came into data world through the data democratization lens and After working in that space. For a long time i found myself serve disappointed by the publicly available tools out there and in the public policy setting world and so became interested in public policy. Eventually after many years. Got my degree in public policy Where i sort of learned about the sort of social economic political and legal frameworks um for use of data and technology. I brought that to the national laboratory system where i worked across the national labs Developing technology information security and privacy policy. And then i took that serve complex bureaucratic experience in brought it to on the city and county of san francisco as the first states first chief data officer and really married my early work in data democratization with What had would emerge to be the open data movement and sort of took not sort of took over the city's on data program overhauled that And then also developed a strategy in executed on it to improve use of data and decision making which eventually led to rolling out a program called data science sf which was on data science as a service using advanced statistical modeling in machine learning to answer questions that departments cared about.

kathleen mulch cog melika today melinda
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

01:48 min | 1 year ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"Hello and welcome to the today podcast. I'm your host Kathleen Mulch and I'm your host Monash Malsor our guest today is shift. Mishra. Who is the head of Medicare retention analytics at CVS health high shift. Thank you so much for joining us on. Today. Absolutely my pleasure. Thank you for having me on August. Yet thanks so much for joining us today. We'd like to start by having you introduce yourself to our listeners. Tell them a little bit about your background because I know it's a very rich background. I WanNa make sure that that they.

Monash Malsor Kathleen Mulch
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:42 min | 1 year ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"Hello and welcome to the AI Today podcast. I'm your host Kathleen Mulch. And I'm your host bottled schmelzer Our Guest today is Kelly fellow who is the director of core data science at the Home Depot Hai Khalifa. Thank you so much for joining us on AI today. Hi guys. Thanks for having me. It's my pleasure. Yeah, welcome Khalifa and thanks so much for joining us. We'd like to start by having you introduce yourself to our listeners. Tell them a little bit about your background and she current role at the Home Depot. Sure. So my name is Kelly fell Jetta. I have PhD degree in computer science. I started my career in data science back in June 2013 as a PhD intern at Careerbuilder, which is one of the largest job boards in the US and the my career with Career Builder actually took extended to until 2018 during that I was actually leading the search and recommendation data science team where I was lucky actually need to get involved early enough and building the semantic search engine for the company and after that building an AI based recommendation engine dead. So the semantic search engine actually is the one that has been leveraged by the company for their be to be sort of business and the day I guess recommendation engine which we built their home is now serving millions of job-seekers on the BTC side of the company. So very proud of that Journey with Career Builder in 2018. I joined Home Depot and I joined as a senior manager, of course recommendation data science team under the online business of Home Depot, I build the team and we actually worked very hard and the last two years to build again state-of-the-art e-commerce recommendation engine for Home Depot, very proud of what we accomplished as a team found in May this year twenty-twenty. I was promoted to director of course data science in my organization. Now, I have the court search data science team called recommendation wage. Science team and the visual AI team our focus our my route Focus now is as the name suggests to improve the core functionality of homedepot.com home from search and documentation perspective. So we work to improve sexual even see we work to make our recommendation more and more personalized and relevant to our customers and guide our customers and kind of give them the experience which they get in the physical store as part of our interconnected experience initiative. So that's overall. What am I roll includes now at Home Depot and I'm very proud and excited actually about the team that we have built for the core data science at Home Depot on the work that we have done that for the e-commerce, you know, that's that's fantastic. And you know, I I really have to give a plug for the talk that you gave at the data for a i week online conference because you you showed you age. And about thirty forty minutes really walking in Fairly good detail how the Home Depot actually does its product recommendation system. We showed how the system works. There was some math in there, which is great all the time a little bit of code more math than code showing how it was the song and it was fantastic. I mean and so, you know for those who are listening if you really wanted to to dive deeper and see this the presentation you can the the conference is available for free. So if you go to data a icon did a i c o n f c o n f, and look for a Khalifa's presentation page, it's on the e-commerce system and talks about the recommendation system. It's just fantastic and I love seeing it because you know, I have to say I'm you know, probably like many of us here in the United States now have a big Home Depot customer feel. I feel like I go there like every other week, especially, you know, we're all at home these days so you can't help but notice the things that you need to write a fix and repair right and they even do some stuff outside job. And it's it's it's the season of the deer kind of eating everything and Wrecking everything. So so I think it's fantastic what maybe maybe for our listeners here? If you can provide a little bit of insight you talked a little bit about the recommendation system. I know that it's really hard to we don't have slides here on a podcast that's going to be hard to share. But you were talking about solving challenging e-commerce problems using the power of data science as a Todd the title of the talk. So maybe you can share some of the insights that you shared at the conference around the recommendation system round recommendation systems in general maybe around the relationship between data science and e-commerce, which you know, maybe people haven't thought about that deeply Yeah, sure sure. And first of all, thank you for highlighting the talk. Absolutely. It was actually a great conference overall. So I congratulate you guys on the success of the conference just enjoyed being part of it. Thanks for having me back to the question about the talk and the relationship between the e-commerce and and the data science absolutely data size is transforming retail to the boss really on the e-commerce side and how we do things and the e-commerce and they use cases I presented in my talk. We're actually real use cases of things that we implemented at Home Depot on faith and that changed actually How We Do recommendation on our websites to make them more relevant and to make them as they mentioned earlier and more personalized to our customers need. So

AI Khalifa Kelly Home Depot Kathleen Mulch Career Builder US Trends Technologies intern director Careerbuilder
Interview with Khalifeh Al Jadda, Director of Core Data Science at The Home Depot

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:42 min | 1 year ago

Interview with Khalifeh Al Jadda, Director of Core Data Science at The Home Depot

"Hello and welcome to the AI Today podcast. I'm your host Kathleen Mulch. And I'm your host bottled schmelzer Our Guest today is Kelly fellow who is the director of core data science at the Home Depot Hai Khalifa. Thank you so much for joining us on AI today. Hi guys. Thanks for having me. It's my pleasure. Yeah, welcome Khalifa and thanks so much for joining us. We'd like to start by having you introduce yourself to our listeners. Tell them a little bit about your background and she current role at the Home Depot. Sure. So my name is Kelly fell Jetta. I have PhD degree in computer science. I started my career in data science back in June 2013 as a PhD intern at Careerbuilder, which is one of the largest job boards in the US and the my career with Career Builder actually took extended to until 2018 during that I was actually leading the search and recommendation data science team where I was lucky actually need to get involved early enough and building the semantic search engine for the company and after that building an AI based recommendation engine dead. So the semantic search engine actually is the one that has been leveraged by the company for their be to be sort of business and the day I guess recommendation engine which we built their home is now serving millions of job-seekers on the BTC side of the company. So very proud of that Journey with Career Builder in 2018. I joined Home Depot and I joined as a senior manager, of course recommendation data science team under the online business of Home Depot, I build the team and we actually worked very hard and the last two years to build again state-of-the-art e-commerce recommendation engine for Home Depot, very proud of what we accomplished as a team found in May this year twenty-twenty. I was promoted to director of course data science in my organization. Now, I have the court search data science team called recommendation wage. Science team and the visual AI team our focus our my route Focus now is as the name suggests to improve the core functionality of homedepot.com home from search and documentation perspective. So we work to improve sexual even see we work to make our recommendation more and more personalized and relevant to our customers and guide our customers and kind of give them the experience which they get in the physical store as part of our interconnected experience initiative. So that's overall. What am I roll includes now at Home Depot and I'm very proud and excited actually about the team that we have built for the core data science at Home Depot on the work that we have done that for the e-commerce, you know, that's that's fantastic. And you know, I I really have to give a plug for the talk that you gave at the data for a i week online conference because you you showed you age. And about thirty forty minutes really walking in Fairly good detail how the Home Depot actually does its product recommendation system. We showed how the system works. There was some math in there, which is great all the time a little bit of code more math than code showing how it was the song and it was fantastic. I mean and so, you know for those who are listening if you really wanted to to dive deeper and see this the presentation you can the the conference is available for free. So if you go to data a icon did a i c o n f c o n f, and look for a Khalifa's presentation page, it's on the e-commerce system and talks about the recommendation system. It's just fantastic and I love seeing it because you know, I have to say I'm you know, probably like many of us here in the United States now have a big Home Depot customer feel. I feel like I go there like every other week, especially, you know, we're all at home these days so you can't help but notice the things that you need to write a fix and repair right and they even do some stuff outside job. And it's it's it's the season of the deer kind of eating everything and Wrecking everything. So so I think it's fantastic what maybe maybe for our listeners here? If you can provide a little bit of insight you talked a little bit about the recommendation system. I know that it's really hard to we don't have slides here on a podcast that's going to be hard to share. But you were talking about solving challenging e-commerce problems using the power of data science as a Todd the title of the talk. So maybe you can share some of the insights that you shared at the conference around the recommendation system round recommendation systems in general maybe around the relationship between data science and e-commerce, which you know, maybe people haven't thought about that deeply Yeah, sure sure. And first of all, thank you for highlighting the talk. Absolutely. It was actually a great conference overall. So I congratulate you guys on the success of the conference just enjoyed being part of it. Thanks for having me back to the question about the talk and the relationship between the e-commerce and and the data science absolutely data size is transforming retail to the boss really on the e-commerce side and how we do things and the e-commerce and they use cases I presented in my talk. We're actually real use cases of things that we implemented at Home Depot on faith and that changed actually How We Do recommendation on our websites to make them more relevant and to make them as they mentioned earlier and more personalized to our customers need. So

Home Depot AI Khalifa Kelly Director United States Career Builder Kathleen Mulch Intern Careerbuilder Senior Manager Todd
Interview with Carlos Rivero, Chief Data Officer for the Commonwealth of Virginia

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

06:25 min | 1 year ago

Interview with Carlos Rivero, Chief Data Officer for the Commonwealth of Virginia

"Hello and welcome to the AI Today podcast. I'm your host Kathleen Mulch. And I'm your host Ronald schmelzer Our Guest today is Carlos Rivera. Who is the chief data officer for the Commonwealth of Virginia off Carlos. Thank you so much for joining us on AI today Hey Ron. Thanks for having me. Yeah, welcome Carlos and thanks so much for joining us. We'd like to start by having you introduce yourself to our listeners. Tell them a little bit about your background. Check your current role as Chief data officer. Fantastic Kathleen. So yes in my current role on the chief data officer for the Commonwealth of Virginia before that. I've been in that role since August of 2018. And before that I was a chief data officer and chief Enterprise architect or the Federal Transit Administration at the US Department of Transportation that was there for a little over two years as well. And then prior to that I was physical scientist with Genoa Fisheries down at the southeast Fisheries Science Center for about fifteen years. So I've been in public service right now going over nineteen years in both federal and state experience. Well, that's great. I think that provides a lot of real Nice diverse set of experience, you know from Fisheries to the federal government to state government. And I think that's part of reason why we love to have your participation that we had your participation at the data for a-week confirmation that ran from September 14th thru 18th 2020 was of course a virtual conferences everything as these days and we were focusing on the data side of AI and for our listeners who may be interested that content is actually still available so you can come and you can hear the panel that Carlos was on when we were focusing on some of the state and the local challenges for AI and data management. If you go to data, that's spelled like data package i c o n f. It's free so you can go on there and you can check all that content will be made available for many months. So you definitely should check it out and Carlos was on a panel really sharing some of the unique insights of applying a machine learning and also some of the interesting challenges of wrangling data at the state level. So maybe Carlos you for those who weren't Intense or maybe even to motivate folks to listen to the family. What are some of them? Sites that you have seen in terms of just the challenge of managing data and getting it to do some magical things like machine learning at the state level. Well, I mean really one of the most basic things is getting people involved in the process. And in fact has plays a key role in that obviously more, you know, as we kind of evolved in once a leveraged data as of CJ Cassat within the Commonwealth, we realize that the participation of individuals not just horizontal across the organization, but also a vertically through different levels of state government is critical for our ability to integrate those data assets in a meaningful way and when I talk about the vertical, how are the patients I'm talking about, you know data storage data custodians data owners executive sponsors being able to participate in the overall governance discussion because everyone has a role to play in our ability to leverage data as a CJ asset to be able to incorporate that into our data analytics to write better intelligence and within that, you know, a comes in machine learning and artificial intelligence briefing. Jane as much value and insight from the data assets than we currently have. Yes indeed. Go ahead Kelly. Yeah, definitely and kind of to follow up with that on this podcast. We talked a lot about Ai and data at the national level, but maybe you can dig a little bit deeper into what are some of the unique challenges around data at this point level because I know that you know in general there's a general data challenges, but then we can also talk about you know, there's differences between State versus local versus Federal. So the fun thing about state is that you get to play with all the businesses at one time, you know in the federal space like when I was no Fisheries, we're very focused on fisheries and Fisheries applications. Mind you, you know as a physical scientist. I really worked with a lot of different data sets. As I was really more in a fraction of those individual populations and their environments right and anthropogenic impact on those environments and how does that change the behavior of individuals within a species right? And so you have to look at the bigger picture and yep. Integrate data from a variety of different sources other Noah Services resolved as live in North Fisheries, but we also have satellite service. We have the ocean service. We have the weather service. So being able to bring in data assets from a variety of different Services different lines of business. If you will to give you a better picture of what's happening in an environment that's very unique like more often than not individuals within that particular industry. We only focus on the data that they collect they work with on a regular basis and not really look at the bigger picture of what other data assets they can bring in same thing for in Federal Transit right in Federal Transit. It was very limited in their you know, what their perspective was with regards to you know, what data asked us what we going to bring in to really understand what's happening out in the world. They're really focused on providing, you know grants of Transit agencies and authorities to make sure that people are able to get to use public transportation in the most effective way. So it's very very silent. But then when you talk about a state government, can you talk about you know being able to leverage data as an asset at that level you really talking about across all of the different page? Business whether it's education Transportation criminal justice, you know environment what-have-you Health, you know, all of those lines of business now come under your purview and you really have to start to understand. What are they unique perspectives and how can you engage those individuals within each of those lines of businesses to be able to see the value in integrating their data assets and making better data-driven decisions home from that integration. So from a state perspective you really start to get a better handle on the overall picture of what's happening out in the real world versus a very I don't want to use this term negative in my topic view of you know, what your assembly looks like and only that which Falls but then you're suddenly are you paying attention to but at the same time, I've also realize that you know data governance and use of data as an ass is really a fractal type of problem where it doesn't matter. What kind of scale you look at it. It's going to have the same patterns associated with some of the same issues that we dealt with at the federal level we deal with birth. Level we deal with at the local level because it's not a matter of our these issues different. It's just a scale at which we operate in that just kind of gives you a little bit of a difference in wage issue is but the reality is that it's very poor the majority of the issues we do with with regards to data governance and data sharing and leveraging data and analytics a machine learning really comes back to the process and the people aspect of the peace process technology interaction.

Chief Data Officer Carlos Rivera AI Fisheries Kathleen Mulch Virginia Scientist Federal Transit Administration Genoa Fisheries Southeast Fisheries Science Ce Ronald Schmelzer North Fisheries Cj Cassat Officer Us Department Of Transportatio Chief Enterprise Architect Jane Kelly
Interview with Shyamala Prayaga

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

06:46 min | 1 year ago

Interview with Shyamala Prayaga

"Hello and welcome to the today podcast. I'm your host, Kathleen Mulch and I'm your host Ronald smells our our guest. Today is Shammala Pro Yoga who is the autonomous digital assistant vision lead at Ford High Shinola. Thank you so much for joining us today on today. Thank you so much. It's my pleasure. Yes. Welcome Chamois, and thanks for joining us. We'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background and your current role at Ford. So. Yeah. I'M GONNA and I'm the frontal known job. You know your aren't complete. So basically, I lead the wishing funded autonomous official assistance the now Newton values cases in what technologies do we need, what kind of experience we need to design, and then bringing it onto the order to find the expedience exempted. So I've been with old saw announced deals. Now on food I will voice box technologies, which is now at all students. On, board with Amazon and that is weird. Mijo. Needed on I will with little up signs once size companies as well. So I've been in the expedience design roller far on mostly two decades now today's about me. Yeah. Well, that's great. Well, you know we do spend a lot of time talking about all these different applications of AI. We call the seven patterns of Ai and we were really excited to have you present. And participate in a panel at our data for a conference which was held live September fourteenth through eighteen, twenty, twenty virtually of course, because everything is virtual but we had you WANNA panel and it was fantastic new shed some really great insights about a especially as it relates to voice assistance and voice and autonomous systems for data, and for those who are listening if you weren't able to attend the event have no worries because. You can still go on and you can access all the content at Data Ai C., O. N. F. DOT COM including chamois panel. So maybe as a preview to encourage those to listen to the whole panel. Chamois, why don't you tell us a little bit about your insights that you have gleaned about using I and the data challenges and some of the other challenges especially in the context of Autonomous Invoice assistance? Yeah. So so basically Voice as. You donate anyone using voices June board noticed agile. The superficial level of poison system is really good. You know when you ask the the assistant to do something, it would answer right. But Dan if you look, there's lot of technologies a lot of different kinds of things going in the background, which kind of makes the assistant waters doing fudd exam. You said play music, veteran recognizing usage music, and understanding the Indian that you meant you want to listen to music who processing eight and then playing the music for you, and then you know natives funding backfield wrist on of these things, acknowledges, and of course, throughout this fish cutting your wise to understand. It to renege does not of data required to understand the national and then of course, you not buying your accounting guide. So there's none of the Dow which needs to be captured throughout when you're designing. Now, I also spoke about like how these voice assistance are limited to. Happy. Use Case is solo voice assistance will officially if you are you know on American with perfect English. But if you know the moment you start having some sort of accident, you know they not as much our diamond, not even the ignites if you how strong accident. So that's another problem. So of course, this dude that challenge they've been handed down from uh specific segment, not the other. Thing especially in the autonomous eighty taints they if you on an automated states is not of background nice but you cannot control. So in those kinds of scenarios, of course, in all the assistance failed ignites, but the users said. So some of the automotive companies they would say instrument the use of Leixoes the window like make sure that business background noise. But then those are not the solutions to simplify the experience of making us Seinfield behalf not of. Their time beyond making improvements everyday. But I still feel like there's more data which is required make expedience use of the US, and that is where you know we spoke about takes and how Imboden is because as we start collecting on all these data than we are getting in, do the tribal seeded as trying to be on those kinds of face? Yeah. You know that's great. I enjoyed that panel so much on the panel we also discussed humanizing privacy Can. You share with our listeners, what this means and Wyatt? So important yeah. Glad you asked him that question also writing a book about this topic, which will come in twenty twenty, one call humanizing privacy so like I mentioned. It takes his on about principle integrity fairness and responsibility dry, and then you know into would be as will. But if you look at these assistance, the biggest challenges you know are great. But then the moment that comes to do is not of issues. For example, you know like it has been so many news at Alex saw Google has been regard late listening to other things. Other reasons they were not supposed to not for pressing for the other reason. So they will those kind of leaks which has happened and even if. You know like when you first buy your device on renewables `integrated. Donate, adopt your device. You would have Johnson conditions and everything in place but then who reads the domes and conditions to know Lakewood did I actually, what do you are starting on what they are using to fulfill specifically quiz no-one really does and that is where the biggest thing is although the companies are trying to cover legally they are covering in a way which is not used as human center and I believe in Ruin ising rival. See because I feel like privacy is the stepping stone towards trust if used those feel lake, this company has the right kind of takes the ad being responsible being the being honest. The Indy Vega behalf dignity than of course, in all people will want to use it in how we do that. So I believe that it's not just the responsibility of legal or you know alike general counsel to design some Johnson conditions concerns on these the. But I believe that business something where you know that it has to be more things we need to be humanizing the entire thing privacy as fake.

Johnson Ford High Shinola Kathleen Mulch Mijo Ronald Ford United States Amazon Official Lakewood General Counsel DAN Seinfield Fudd Wyatt Google Imboden Alex
Interview with Shyamala Prayaga

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

06:46 min | 1 year ago

Interview with Shyamala Prayaga

"Hello and welcome to the today podcast. I'm your host, Kathleen Mulch and I'm your host Ronald smells our our guest. Today is Shammala Pro Yoga who is the autonomous digital assistant vision lead at Ford High Shinola. Thank you so much for joining us today on today. Thank you so much. It's my pleasure. Yes. Welcome Chamois, and thanks for joining us. We'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background and your current role at Ford. So. Yeah. I'M GONNA and I'm the frontal known job. You know your aren't complete. So basically, I lead the wishing funded autonomous official assistance the now Newton values cases in what technologies do we need, what kind of experience we need to design, and then bringing it onto the order to find the expedience exempted. So I've been with old saw announced deals. Now on food I will voice box technologies, which is now at all students. On, board with Amazon and that is weird. Mijo. Needed on I will with little up signs once size companies as well. So I've been in the expedience design roller far on mostly two decades now today's about me. Yeah. Well, that's great. Well, you know we do spend a lot of time talking about all these different applications of AI. We call the seven patterns of Ai and we were really excited to have you present. And participate in a panel at our data for a conference which was held live September fourteenth through eighteen, twenty, twenty virtually of course, because everything is virtual but we had you WANNA panel and it was fantastic new shed some really great insights about a especially as it relates to voice assistance and voice and autonomous systems for data, and for those who are listening if you weren't able to attend the event have no worries because. You can still go on and you can access all the content at Data Ai C., O. N. F. DOT COM including chamois panel. So maybe as a preview to encourage those to listen to the whole panel. Chamois, why don't you tell us a little bit about your insights that you have gleaned about using I and the data challenges and some of the other challenges especially in the context of Autonomous Invoice assistance? Yeah. So so basically Voice as. You donate anyone using voices June board noticed agile. The superficial level of poison system is really good. You know when you ask the the assistant to do something, it would answer right. But Dan if you look, there's lot of technologies a lot of different kinds of things going in the background, which kind of makes the assistant waters doing fudd exam. You said play music, veteran recognizing usage music, and understanding the Indian that you meant you want to listen to music who processing eight and then playing the music for you, and then you know natives funding backfield wrist on of these things, acknowledges, and of course, throughout this fish cutting your wise to understand. It to renege does not of data required to understand the national and then of course, you not buying your accounting guide. So there's none of the Dow which needs to be captured throughout when you're designing. Now, I also spoke about like how these voice assistance are limited to. Happy. Use Case is solo voice assistance will officially if you are you know on American with perfect English. But if you know the moment you start having some sort of accident, you know they not as much our diamond, not even the ignites if you how strong accident. So that's another problem. So of course, this dude that challenge they've been handed down from uh specific segment, not the other. Thing especially in the autonomous eighty taints they if you on an automated states is not of background nice but you cannot control. So in those kinds of scenarios, of course, in all the assistance failed ignites, but the users said. So some of the automotive companies they would say instrument the use of Leixoes the window like make sure that business background noise. But then those are not the solutions to simplify the experience of making us Seinfield behalf not of. Their time beyond making improvements everyday. But I still feel like there's more data which is required make expedience use of the US, and that is where you know we spoke about takes and how Imboden is because as we start collecting on all these data than we are getting in, do the tribal seeded as trying to be on those kinds of face? Yeah. You know that's great. I enjoyed that panel so much on the panel we also discussed humanizing privacy Can. You share with our listeners, what this means and Wyatt? So important yeah. Glad you asked him that question also writing a book about this topic, which will come in twenty twenty, one call humanizing privacy so like I mentioned. It takes his on about principle integrity fairness and responsibility dry, and then you know into would be as will. But if you look at these assistance, the biggest challenges you know are great. But then the moment that comes to do is not of issues. For example, you know like it has been so many news at Alex saw Google has been regard late listening to other things. Other reasons they were not supposed to not for pressing for the other reason. So they will those kind of leaks which has happened and even if. You know like when you first buy your device on renewables `integrated. Donate, adopt your device. You would have Johnson conditions and everything in place but then who reads the domes and conditions to know Lakewood did I actually, what do you are starting on what they are using to fulfill specifically quiz no-one really does and that is where the biggest thing is although the companies are trying to cover legally they are covering in a way which is not used as human center and I believe in Ruin ising rival. See because I feel like privacy is the stepping stone towards trust if used those feel lake, this company has the right kind of takes the ad being responsible being the being honest. The Indy Vega behalf dignity than of course, in all people will want to use it in how we do that. So I believe that it's not just the responsibility of legal or you know alike general counsel to design some Johnson conditions concerns on these the. But I believe that business something where you know that it has to be more things we need to be humanizing the entire thing privacy as fake.

Johnson Ford High Shinola Kathleen Mulch Mijo Ronald Ford United States Amazon Official Lakewood General Counsel DAN Seinfield Fudd Wyatt Google Imboden Alex
How AI is Being Applied to the Stock Market

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:23 min | 1 year ago

How AI is Being Applied to the Stock Market

"Hello and welcome to the Today podcast I'm your House Kathleen Mulch. I'm your host Ronald Schmeltzer. So today we're going to spend our. podcast talking about our use case series where we look at the adoption, I in various different industries and today, we're GonNa take a look at the stock market and the training on. Amazing. Other things like the stock market commodities and futures bonds and Blah, Blah Blah. So we talk about stark record. We're talking about the trading floor in general, but basically the stock market and it is the dream. It is the long dream of anybody WHO's in the financial trading industry to find the magical algorithm magical model that can predict the future of the market because if you can do that you basically when everything in life. Right because. If I can tell you the price of Amazon or apple stock or oil, or whatever tomorrow, I can lock in my buy low sell high sell high by. Then you went and you know what? There are very smart PhD mathematicians out there right now, who are working on machine learning models to predict the markets, right? So we should talk about. I think ever since there's been a market, there's been someone trying to figure it out in one way or another. Just. Technology helps but still hasn't cracked it. So like Ron said, we wanted to spend some time today talking about a and how it's impacting the overall stock market trading floor and that whole area recently adoption of AI in financial trading has seen an uptick and more people are starting to look into this wealth managers are using ai to help serve their clients better. Freighters are using AI and also Also, augmented intelligence tools to gain insights and flight market advantages. You know anything that they can get. They really try any slight gain, is worth it to them, and also many people are starting to see some real value from using ai just in general. So we wanted to spend some time to dig deeper into this and talk about how ai is being applied in various ways. So part of getting. Getting an advantage in the stock market, of course, was called Informational advantage. If you have to know something before other people know things. Then you have an advantage just goes all the way back to you. Even Barron Rothschild in London stock markets, how he used the supposedly passenger pigeons ticket information from the battlefields. You know that was like supposedly like one of those big innovations and it kind of carries. Carries through to the high frequency trading. We're like you can engage in micro. Senate? Like not even Microsoft, by Pico Senate seconds of advantage over someone else because you have the man, you could place that trade just before the other people place the trade. Then you have an advantage, right? So having information advantage is a real thing, but you know nobody really knows the future especially machines. So really. So what you can use, you could try to just build better models. You could try to figure out, could try to fill the basically figure out a better cause and effect like when you see something happening in one place, right? What does your model tell you about what will happen somewhere else or maybe something in the same market like when I see, let's see there's oil prices are going up does your Mama. Mama predict something with some high rate of predictability about what will happen in other market, and so this idea of modeling model changes for Stock Market for finance for risk levels, the is really able to really analyze things just a much higher level. It's basically just doing things that skill we talk about that with the idea of hyper personalization is one of the patterns just in general patterns and anomalies and predictive analytics. Systems can read all the news, they can look at all the social media. Posts, you know they could check every stock, it check derivatives on stock options that could check all those things. So companies that are really using a I are using it first and foremost to just be a massive intelligence gathering tool and synthesize all that information together, and basically try to use that to inform and influence their models and use those models to basically help do things at a much faster speed, and of course, is about speed right? The speed is incredibly valuable and so someone getting something just slightly faster getting that insight. Maybe you just haven't inside that nobody else has which is great in. You can act on that and hope nobody else notices, but a lot of time. But the truth of the matter is, is that you never alone there's too many people looking at too many things. So it's really about speed right and there are many many companies that are really building these ai tools to look at the complex patterns in the market and analyze those pens so that that's one big way is hoping right now.

AI Kathleen Mulch Ronald Schmeltzer Senate Barron Rothschild Mama Microsoft Pico Senate London Amazon RON Apple
Interview with Igor Perisic

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

06:19 min | 2 years ago

Interview with Igor Perisic

"Hello and welcome to the today podcast. I'm your host Kathleen Mulch and I'm your host Ronald Schmeltzer. Our guest today. It's ego or parasitic. Who is the chief data officer and VP of Engineering at Lincoln? High your thank you so much for joining us today. Yeah thank you eager for joining us. We'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background and your current role at Lincoln certainly I guess from an education perspective. I grew up in Switzerland. And I have an Undergrad degree in applied math from the which are then follow up with graduate studies in statistics in the US somebody A statistician and they not only my career just pivoted to the industry and since then was essentially involve which was originally called it say applied statistics than data mining machine learning and data signs. And now they I. I guess it's GonNa depending on what it was at that time to liberal. Just change the funny thank you though is at the beginning. When I studied into domain. It didn't have the labels and people didn't understand what it was interested in doing or what I was doing so they would always people back to statistics until you're doing stuff like the Census Bureau or FDA drug pools etc etc. Any took awhile to sorta registered in the sense from the beginning in my career. I was always interested in to leveraging data provide to enhance individuals ability to do the job and to achieve more today Lincoln. I'm the chief data officer and the VP of engineering part of the engineering orgnization. From the beginning. Because I'd like to build things and the my responsibility than on the engineering side. It's everything that covers the data spectrum from online to offline distributed systems that at all as to store data about storage systems with semantics so the data flow from one service to another one thinks like Cobb we invented on open source. I think to other big systems at At scale think ado spark etc etc and then on top of infrastructure and these might teams are responsible for all the ai that enables us to percents experience so mentors and on the chief data officer side responsible to make sure that we use data in responsible fashioned according to the terms of the service to regulations set kind of city between product legal and engineering to make sure that we on the same page and inflammation communication pros across so the everybody's consulted than making sure. That knows what's happening your excellent well. You know the interesting. You mentioned data sciences sort of both the profession and sort of as a role has really emerged quite a bit most very recently near. Right. It's interesting people. You know when you look at so trying to understand a little data science when one. You like really isn't very much You'd think there'd be a well established definition menu. Some people look at it from the statistics probability perspective primarily or data analytics perspective. Some people look at it from the data management perspective and say mostly data and a little bit of science and some people. So it's kind of interesting all that but the chief data officer role also is fairly new thing. But you know of course. We've been dependent on data for decades. You know ever since we've had data that also changes through time to as data. A science at the beginning was doing a lot of things which today you we already saw the state. For example. A data engineer was co composed science at the beginning. Like around two thousand seven eight just because all these data systems did not exist so you have to build them. And there's the need they'd be innovation to dry them and who felt that need while the data scientists so they did engineering was part of it. Same thing she did officer Demand on making sure that the way that individual manipulate or use data within their organization is much more controlled than what he was before kind of elevated that will people flocking to download from different perspectives. Yeah people that come from the legal side and then got these towards engineering. People would come from. They did assign side More to woods through legalese as well as engineering saddened some people will come from the engineering side. Any understand what the legal component than the science components. I've always changes through time. The label kind of stays but the definition of what you really doing. Changes Adopts Yeah. I think that's important and we actually do like the idea of data centrisly understanding date on the role of data at the C. level which means that has visibility at the highest most strategic level there orgainzation traditionally maybe people have thought of as a component of the officers role the technology officers role. So that's really good to hear your perspective on especially at Lincoln so a one of the things people may not be aware of. You know. Obviously one of the things. We're going to try to do it this. Data's trying to get more insights and and use data to train systems for machine learning to apply to a wide range of applications. And so you know. Lincoln in particular uses a in many ways that maybe perhaps users may not be aware of on a daily basis. So can you give us some sort of insights and outline of the many different ways that linked in is applying? Ai and how it's enhancing the user experience so we leverage ai at Lincoln always in the context of our vision of using us to create economic opportunity for every member of the global workforce and within that space we described as being the oxygen for Linton. It's embedded just about everywhere you think about creating economic opportunity at the scale that want to create it. We have roughly six hundred seventy five million members on the platform. There's roughly nowadays. I'm not too sure. How many jobs anymore but As well doing that match between that big of a set of individuals and that big of a set of jobs you need to have tools you need to have a to allow it to be personalized to be good on that level so you have a lot of things that we do is to making sure that hold that information is personalized to make individuals chronicled smarter and enable them to do their work better but that did not flooded with the noise it could be in the patron. That's on the visual. Side is also very big pot which is Behind the scene for example we do. Now he'd be testing platform the way that we see it out how to root out traffic to all this end to these access points turntable. Speedy delivery experienced members was simple things of simple uncomplicated. Things Anomaly detection services would something is different than what? We focused sweetie just about everywhere.

Officer Lincoln Vp Of Engineering Kathleen Mulch Ronald Schmeltzer Switzerland Census Bureau United States AI Cobb Engineer FDA
ML Ops in Practice: Interview with Seth Clark, Modzy

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:50 min | 2 years ago

ML Ops in Practice: Interview with Seth Clark, Modzy

"Hello and welcome to the AI today. Podcast I'm your host Kathleen Mulch and I'm your host Ronald Schmeltzer our guest today South Clark. Who is the head of Product at Monte? Hi Seth thank you so much for joining us on today. I'm really excited to be here. Ron Kathleen looking forward to a great chat. Yeah we're really excited to have you too. So Seth we'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background and some exciting things that are going on at Monte absolutely show I come to through a somewhat meandering pass run and actually share an almond butter both engineers over MIT. And I was like to say I learned the arts of withstanding torture really well for my time there which is a streaming while to the course of my career. I don't know if you had the same vibe when you were going to put you through the ringer but you know if that's good. Keep everybody on their. That's right so I actually which meant now as I studied ocean engineering there and then I ended up actually getting my master's degree in yacht design. Wow so you might wonder. How does someone with kind of engineering background? The time is in a lot of computational fluid dynamics and up in this space long circuitous path certainly but discover the power of using computers to do math transition and to some modeling and simulation work over the years and found this kinda weird niche in between the world of Software Development Data Science. The wasn't really call that quite at the time that has emerged into enterprise. Ai So from hard engineering background. I kind of found my way into this software space which has been really fun one of the things. I really enjoy it when you're building a physical system design and plan things years in advance before you see something come to life and with software with artificial intelligence and machine learning models you can see something. Come to life in a matter of hours. Which is a really exciting place to be if you're curious. Adhd kind of person. Like I am so through those transitions of the course of my career. I've found myself in a really fun position where I serve as the head of product over Mazi wishes in Ai Platform that we've been developing to really help. Large organizations get a better handle on how they can scale artificial intelligence across some really complicated environments. Some of the things we're looking to do include creating a library of machine learning algorithms. Ai Models you can call them do all kinds of crazy stuff analyzing satellite imagery crossing audiotext taxed and translating audio files from one language to another doing text translation across no millions and millions of the gigabytes of data whole range of different capabilities through this marketplace. Our customers can get access to in also this machine learning operations or MLS pipeline capability meant to help our customers really get an opportunity to have a better handle on how they're implementing artificial intelligence scaling. It securing it in governing it across the enterprise. So it's a pretty fun time to being A and we're having a good time of it over at Monte. Yeah definitely yeah I mean. This is the resurgence of AI. And not my undergraduate adviser was Rodney Brooks. Who really made his name through the second wave of AI? That sort of the robotics wave and the systems wave of late seventies or expert systems wave and early eighties. And of course I kind of came and went but now we're hearing the third wave the summer then. People have discovered rediscovered neural nets and deep learning and of course through the power of big data. And some pretty good computing infrastructure. Where making this stuff work. Where are we couldn't and you know one of those big changes that we've profiled here at cognreznick. As part of our research we track about six thousand minutes the market we'd do about forty or fifty reports year and one of our more. Recent reports was on. These machine learning OPS OPS operations management which is dealing with the fact that when you've built a model that's all nice but things are complicated in the real world right models change and data's changed and we have to manage this model and some people should use the modern. Some people shouldn't use the model and yet the discover the Mommy got security and you got to do all these things track into track changes over time especially depending on the model and so this is the space is fairly new. Even though is like sixty seventy years old you know. It hasn't really been part of the enterprise vernacular until like the last few years so this whole area of Milan is actually really very nascent. So from your perspective. What do you see as the main aspects and components of you know? I think it's the transition from theoretical to applied is really where we discover the need for machine learning operations Raimondo APPs. The theory has been going strong. Ever since the emergence of truly deep neural network architectures back in two thousand and six one. We saw kind of the first broadscale application of Alex but broadly speaking the theoretical side of things. He's been doing a lot of cool stuff as soon as you try. And take all this theoretical work and apply it into a real production system all of a sudden you realize all the gaps that exist when it comes to taking a model that works someone laptop and worked on Lab. Data performed inferences at a really low speed. And now you're trying to pump your entire enterprise's data pipeline through these models sudden. There's a whole lot more that you have to do to get this working right so I think is kind of the the summary of needs the big enterprises and big organizations sort of discovered collectively that need to be met for A. I'd actually return some sort of benefit. I mean investment. That's made into it so that it includes making things more repeatable. Kind of breaking out of his mindset of Monolithic applications adding new security protocols and parameters. You know managing hardware costs because those can be really high when you start doing this at scale so yeah I really see being kind of around application of. Ai

AI Head Of Product Seth Ron Kathleen Kathleen Mulch Monte Adhd Rodney Brooks Ai Platform Ronald Schmeltzer MIT South Clark Cognreznick A.
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

08:28 min | 2 years ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"The today podcast produced by militia cuts through the hype and noise to identify. What is really happening now? In the world of artificial intelligence learn about emerging trends technologies and use cases from called militia analysts and guests experts. Hello and welcome to the AI. Today podcast this is actually a specially. Today podcast episode. I'm your host Ronald Schmeltzer and I'm your host Kathleen mulch today. We're GONNA be talking about how. Ai Is helping to fight global pandemics as you can imagine we are in a global pandemic right now and so we wanted to spend some time on this special podcast talking about how AI is helping to detect co vid fight coated help promote workers and then also how it can be used to help predict the economic impact that this is going to have been paying attention to the news or maybe you're actually listening to this podcast probably a year from now. It's possible people. Listening to our podcast. Even way back to the twenty seventeen will where we are right now. Here when we're recording this podcast in two thousand twenty. We're in the midst of a global pandemic that has really impacted the whole world. You know starting in China. The Corona virus this latest iteration of the Krona virus called Kovic. Nineteen is a strand of illness that has been known that's particularly infectious and has a fairly high mortality rate. Because it's a respiratory illness it's related to other respiratory illnesses like Moore's which is the Middle East respiratory syndrome and SARS severe acute respiratory syndrome. Usually we've been dealing with these sorts of epidemics that have happened in regional places and sort of dealt with them as part of our general healthcare system but this time around it has impacted the world and it's traveling at such a high rate that it's something the whole world has paid attention to and we've dealt with it with some pretty significant measures right everything from you know you can't leave your house and you'd be no social distancing and all these things that are really trying to reduce the impact and spread of this very pernicious virus right because this disease spreads quickly like you said and it's affected basically every single part of the world so right now it has a presence in one hundred and eighty seven countries which is just about every single country in the world and the total number of cases. As of this podcast is rapidly approaching half a million and that's people that have actually been tested and tested for Corona virus. And then there's also been over twenty thousand deaths recorded worldwide as well which is just devastating. So as you can imagine you know. This is very wide spreading fast spreading and it's affecting everybody in every aspect of life and that's why we felt it was important today to have a podcast devoted to this so rahm gave a little bit of background about what is Kovic. Nineteen but how can we detect it? And how can we use? Ai To help detect and treat covert right because at the moment there's no vaccine for Kovic so things that were just trying to mitigate near they're trying to prevent people from spreading this disease and we're trying to find ways to treat it and also there's all these other impacts that Cova having because it's forcing people to work from home so we now have this great experiment where ninety percent of the workforce is a work from home. Of course we've never had remote workers in these quantities but it's also having a tremendous economic impact. You know every person event that's happening right now has been cancelled so no south by South West this year at least so far you know none of these other big events and so people are like moving onto online conferencing. Well that's having another impact on the Internet systems and we're having economic impacts healthcare impacts of course impacts on everything so of course the one place for us to start. I is that is such a transformative technology that it can help with all these factors it can help with the economic the Internet and of course the healthcare impacts and so starting with the most important. Let's focus on how a is actually working to help find and in cure and find treatments for covet and potential cures for treatment. So there's been a lot of interesting activity happening as you can imagine. We have great data scientists who work at these amazing technology companies who have a huge amount of data and tons of compute power. And you can bet that. They are applying that intelligence in those that experience to the challenge of finding treatments for Cova. So we have some examples from. Comey's every announced their various things. We'll tell you about some of the things that have been out and announced so far right so one company deep mind which listeners. I'm sure that you might have heard. Us talk about in previous podcasts. They were acquired by Google. They recently put on the Internet. The sequencing of six protein structures that are linked to Kovic nineteen and. It's too early to verify these results. But it's hopeful that companies like this are working in trying to figure out solutions also. The White House recently put a researchers. They're tasking them to help. Figure this out because this is truly a global pandemic so the United States was a little bit slow on the uptake of Kobe. Nineteen but the White House Office of Science and Technology Policy has urged researchers to employ ai to find solutions to this cova problem. So that's promising that we're hearing the government say you know researchers? Please put effort into this. So as you can imagine things are publishing data sets and their publishing these outputs machine their names sharing it. Which is Great? I mean the Centers for Disease Control. Which is the United States is? This is how we approach. Democ snap and de public health care we have our CDC and every nation has their own CDC and the world healthcare organization well. Who you know. They're basically all pushing researchers to look into this. You know there's twenty nine thousand research documents that have already been produced at the time of this recording and analysis. You know because we got lots of data right. We have data from the hospitals from the treatment facilities from the labs we have tons of medical imaging. We got all this stuff and needs to be scrutinized. The big experiment here. We have tons and tons of data. Now we have real. We don't even need to have synthetic data. We got legit real data from people who need actual treatment and they and for those who from with. Cagle which is sort of a way to share it get data scientists to solve problems. There's a Kagwa cord. Nineteen which is an AI. Challenge that has been pushed online that helps to seek help from interdisciplinary field. So even if you're not in medicine if you're like you know doing predictive analytics for supply chain. You can participate in this challenge to help. Provide input into the data set as part. So this is a way to to speed up this challenge and coincidentally zone addition to this Kathleen and I also host the AI in government event series which is usually held in Washington DC. But just like most event organizers. We've brought it online. And in our march in government event we had two guests from the Centers for Disease Control. Who talked about a crowd? Sourced approach to machine learning so it seemed this approach to basically putting data sets out there. Challenging the general population having them build machine learning models. And then using that rather than trying to build your own internal data team and build your own internal data set and kind of keep everything behind the wall in situations like this where you need to get as much help from as many people this approach to crowdsourcing seems to be getting increasingly more popular. Yeah you know and that was an interesting point that you brought up because some people say yes to crowdsourcing some people say no but in times like these you know you can really get a lot of value from as well so it is really a good approach for also seeing you know we talk about all the time that data is at the heart of ai and that machines are really good at processing large amounts of data very quickly and this is proving handy in situations like this. You Know Ron mentioned that. There's twenty nine thousand research documents that needs to be analyzed and scrutinized. Could you imagine if humans had to go through and do that? I mean that would take forever and we would even farther behind. We're also seeing companies and startups. Who are looking to you know we set that. There's no vaccine for the stroke yet and. I think that that's another reason that people are so scared and uncertain about this. Because there's no vaccine. We can't get immunized for this. So what do we do if we don't have vaccines while we look to other medicines to say? Hey we'll these have similar effects. Can these treat this until we're using?.

AI CDC Kovic Cova Kathleen mulch Ai White House Office of Science United States Ronald Schmeltzer Google Middle East China rahm
Special Episode on How AI is Helping to Fight the Global Pandemic

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:03 min | 2 years ago

Special Episode on How AI is Helping to Fight the Global Pandemic

"Hello and welcome to the AI. Today podcast this is actually a specially. Today podcast episode. I'm your host Ronald Schmeltzer and I'm your host Kathleen mulch today. We're GONNA be talking about how. Ai Is helping to fight global pandemics as you can imagine we are in a global pandemic right now and so we wanted to spend some time on this special podcast talking about how AI is helping to detect co vid fight coated help promote workers and then also how it can be used to help predict the economic impact that this is going to have been paying attention to the news or maybe you're actually listening to this podcast probably a year from now. It's possible people. Listening to our podcast. Even way back to the twenty seventeen will where we are right now. Here when we're recording this podcast in two thousand twenty. We're in the midst of a global pandemic that has really impacted the whole world. You know starting in China. The Corona virus this latest iteration of the Krona virus called Kovic. Nineteen is a strand of illness that has been known that's particularly infectious and has a fairly high mortality rate. Because it's a respiratory illness it's related to other respiratory illnesses like Moore's which is the Middle East respiratory syndrome and SARS severe acute respiratory syndrome. Usually we've been dealing with these sorts of epidemics that have happened in regional places and sort of dealt with them as part of our general healthcare system but this time around it has impacted the world and it's traveling at such a high rate that it's something the whole world has paid attention to and we've dealt with it with some pretty significant measures right everything from you know you can't leave your house and you'd be no social distancing and all these things that are really trying to reduce the impact and spread of this very pernicious virus right because this disease spreads quickly like you said and it's affected basically every single part of the world so right now it has a presence in one hundred and eighty seven countries which is just about every single country in the world and the total number of cases. As of this podcast is rapidly approaching half a million and that's people that have actually been tested and tested for Corona virus. And then there's also been over twenty thousand deaths recorded worldwide as well which is just devastating. So as you can imagine you know. This is very wide spreading fast spreading and it's affecting everybody in every aspect of life and that's why we felt it was important today to have a podcast devoted to this so rahm gave a little bit of background about what is Kovic. Nineteen but how can we detect it? And how can we use? Ai To help detect and treat covert right because at the moment there's no vaccine for Kovic so things that were just trying to mitigate near they're trying to prevent people from spreading this disease and we're trying to find ways to treat it and also there's all these other impacts that Cova having because it's forcing people to work from home so we now have this great experiment where ninety percent of the workforce is a work from home. Of course we've never had remote workers in these quantities but it's also having a tremendous economic impact. You know every person event that's happening right now has been cancelled so no south by South West this year at least so far you know none of these other big events and so people are like moving onto online conferencing. Well that's having another impact on the Internet systems and we're having economic impacts healthcare impacts of course impacts on everything so of course the one place for us to start. I is that is such a transformative technology that it can help with all these factors it can help with the economic the Internet and of course the healthcare impacts and so starting with the most important. Let's focus on how a is actually working to help find and in cure and find treatments for covet and potential cures for treatment. So there's been a lot of interesting activity happening as you can imagine. We have great data scientists who work at these amazing technology companies who have a huge amount of data and tons of compute power. And you can bet that. They are applying that intelligence in those that experience to the challenge of finding treatments for Cova. So we have some examples from. Comey's every announced their various things. We'll tell you about some of the things that have been out and announced so far right so one company deep mind which listeners. I'm sure that you might have heard. Us talk about in previous podcasts. They were acquired by Google. They recently put on the Internet. The sequencing of six protein structures that are linked to Kovic nineteen and. It's too early to verify these results. But it's hopeful that companies like this are working in trying to figure out solutions also. The White House recently put a researchers. They're tasking them to help. Figure this out because this is truly a global pandemic so the United States was a little bit slow on the uptake of Kobe. Nineteen but the White House Office of Science and Technology Policy has urged researchers to employ ai to find solutions to this cova problem. So that's promising that we're hearing the government say you know researchers? Please put effort into this.

AI Kovic Cova White House Office Of Science Ronald Schmeltzer Kathleen Mulch White House Middle East Google Respiratory Syndrome Rahm China United States
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

14:25 min | 2 years ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"The I. Today podcast produced by cowed militia. Cut through the hype and noise to identify what is really happening. Now in the world of artificial intelligence and learn about emerging trends technologies and use cases from politica analysts and guests experts. Hello and welcome to the AI today. PODCAST I'm your host Kathleen Mulch. I'm your host Ronald smells so one of the things we do at cognitive let occurs we produce research on the markets and landscape. We look across all the people that we've spent time talking to the use cases. The case studies all Venice. Did you spend time talking to and we ask ourselves the question once a year. Well how is he. I being adopted worldwide. What's happening in the world with people actually implementing an yeah right right are people or companies are certain regions more heavily adopting ai than others? What's really going on so a few weeks ago? We published the report called Global. Ai Adoption trends and forecast for twenty twenty. So we'll linked to it in the show notes. This one is a free download so we encourage all of our listeners to download it and check out the findings but we're going to spend some time today going through the report itself and highlight some of the key findings and maybe some unique or interesting. MM findings that we found from the survey and the report that we did and so the way that we accomplished this as we serve it we sent out the survey to over fifteen hundred individuals individuals various different companies and countries all over the world and got some two hundred and something responses and use those responses to one inform US specifically what's happening of course allowed allowed to also generalize in some ways about the trends that we're seeing for that if you're interested by the way participating in future surveys. I encourage you to reach out to us. Send us an email to info so I- NFO at Melissa and L. Y.. Dot Com and we would be happy. Include you in future service right so so we'll go through some of the key findings first and then the way that we also broke. This down was by our seven patterns obey because we said okay. Well let's one thing to say people are adopting ai but how truly are. Are you adopting A. Are you doing predictive analytics. Application or chat BOT application or hyper personalization application so some of the key findings that we've found from the the report is that by twenty twenty five over forty percent of the respondents that answered our survey said that they will implement a I in one or more of the identified seven patterns earns of Ai and almost ninety percent said that they'll have some sort of impress a implementation over the next two years. We found those numbers to be. You know very positive positive signs for the industry because it saying almost half will implement one or more pattern by two thousand twenty five so just a few years and then ninety percent so nine out of ten said that they'll have some sort of a m progress within the next two years so you might be thinking. He's looks like contradictory information. Wise that the forty percent or say and do but ninety percent have a project that they're doing two years you would think that ninety percent they're doing well this has to do with of course understanding versus the patterns because we asked them all. Okay yeah great well. Let's not talk about. How many of you are looking at doing a chat Bot and the next two years okay? How many of you are looking at doing a machine learning predictive analytics or recognition project or some sort of automation project? That's using cognitive automation. Oh well all of a sudden now. The numbers started going up and in our chart will show you kind of how the adoption patterns are looking looking because basically when you start looking at the more details yeah machine learning honestly an ai are being embedded in everything and it actually may be difficult to avoid using a machine learning. So even if you're saying well maybe we're not intending to build her on machine learning models it may end up being the end up using them anyways so we also sort of looked looked at turtle how the world was moving with. We're like well. Maybe you know North America Europe bird kind of moving at a different pace Rasiah Africa. Now you know what this is one of the interesting things about I in our research from all respondents and a response come from all over the planet they are all roughly moving at the same pace. It is true that you know Australia. Eliana Asia Europe. They have different timing. What their plans? But basically it's not like we're seeing an over concentration of aggressive plans and North American Europe and less so otherwise it's just it seems like this is just the global movement and then another thing we've talked about process automation a a lot and many companies especially many government agencies here in the. US are very hot and heavy on and process automation. In general. What we found is that fifty four percent of respondents plan to implement a approaches to process automation within the next few years so over half and then fifty two percent of respondents plan into implement a enabled conversational systems by twenty twenty five so those again were not really surprising numbers for us but something that we wanted to point out because because people are finding value in automation and I think in general taking their data cleansing it and then using it for higher level value uh-huh and so when they're able to take cognitive approaches in process automation they're starting to really see value in so we're excited that people who are really moving forward with that? Let me talk about a enabled conversational systems all the time and how companies can use that to help in a variety of different ways that can help with customer service service can also help with. It self service so they could use it internally as well and it's able to allow companies to do more with the same or less resources than before right right so our last sort of key finding sort of digging so some of the more details in a moment is that for the organizations that are sort of struggling with making ai ham or like a haven't quite quite yet taken the step. What they've said is that their biggest barrier to adoption is actually insufficient quantity or quality of data? That's like one of the biggest things things followed by lack of talent so basically people in data for a lot of response to send you know even for the people who are moving ahead. They have acknowledged that these are things that a slowing them down so for the companies that are not planning to implement ai at all within the for the next two years. The thing that they said was the biggest showstopper was just. They haven't yet justify the Arwa. which kind of makes sense or that? There isn't enough of an advantage of AI. Over non approach sprang for the ones that have taken that next episode okay. I think there's I want to do this project you know. I have have an R. Y.. I think it's going to give me an advantage. There getting stuck on people and data right right and that really comes as no surprise because cleaning data it can be a very very manual process very time intensive it can also be very costly as well and depending on the sensitivity of that data. That depends on what vendor or you can go with and where physically the data needs to be cleansed and crapped then followed by limited availability for a talent and skills. We've talked about. There's a big telling cruncher especially around data scientists so some of these smaller organizations. Just don't have the money to afford a data scientist on their team. So what can and they do then. That's where they're limited by hiring talent so digging a little bit deeper in one of the things we did our report is we asked them say okay. Well how many of you doing okay. Great Right now. How many of you are doing hyper civilization or pattern about is a nominally? Is predictive analytics. No automation which is not AI. We spent many reports talking about that. But we do track it because it is that when those pathways to get Aso we talk about process automation separate from autonomous systems. And then we talk about conversational systems recognition recognition systems and then goal driven system. And maybe it might not come as much of a surprise but the thing that's been the most widely implemented as of last year conversational national systems chat bots voice assistance Alexa skills. You know smart tech spots and embed. Yeah because you may not necessarily be thinking of those things is they. They're are all powered by machine learning especially the constitutional system. I you know high rate of adoption moving at a very sort of steady in four grants annual growth that the annual growth but like the OT overall adoption is like twenty percent and twenty four percent just keeps growing that episode. The thing that's kind of more interesting is the ramp. The rate at which people are implementing running process automation right or plan to within the next few years so in twenty nineteen there were about. Ten percent of participants had are bought in production production. Twenty twenty about sixteen percent but by twenty twenty five so five years from now fifty four percent so one out of every two every other. Yeah half half of the people want to have. Rpa In practice and implementation at their various companies and that says a lot to the growth in the potential of that market. No the of course new anytime you asked what predictions where do you think you're going to be in five years. I mean it's Kinda hard to necessarily predict but but the reason why this is. I think is sort of an expectation that it's like yes to the extent that I can use help automating back office processes and the tools are out there and their well developed developed. It's quite not having them so they're not gonNa do it but the the other half said they won't so it's like okay. You know this is kind of a lot of companies is becoming more and more of a no brainer. There's some things that are just it's GonNa take a long time goal driven reinforcement learning. That's seems to be a slow go for a while. Yeah I mean we've presented on various use cases for goal driven systems and you can can use it for resource optimization and stuff but it seems to still lag with adoption of many you know general use escape for us the other one of course is the recognition systems that one is driven by use case. That's mostly like you know. Why do I need to image recognition voice recognition? It's not that it's just a matter. Her of just waiting over time. It's just that it's for that pattern. It looked like it was highly dependent on what the specific use case. There's such a need that so so why does it matter if ten. I almost don't need it right. And then another pattern. That was not heavily. Seeing adoption was the autonomous systems pattern. And now we broke out the process. This automated from autonomous systems. Even though they all technically fall within that bat pattern because we found that when. It's a truly autonomous you know machine learning thank system. That's a very hard problem to solve. We always recommend in our trainings that you do not start with that problem because it is. The goal of the autonomous optimist system is to minimize and greatly reduce or eliminate the human from the process. So that's a really hard problem to solve. That's why the adoption is is not going to be super high. Not all use cases require that and it takes a Lotta time and resources to plan that and make sure that it's really truly accurate Britt. So that's another reason why we saw adoption of that. Not being as high as some of the other patterns so as mentioned earlier we talked about plans for adoption of ice. Let's go back to the general. Are you planning on implementing. I talked about this. The the forty percent versus the ninety percent. They were just talking about them and we asked in general so like in general you know only like a small small number of people also. They had no plans in general. The rest had had multiple plans. And some some now. Maybe that's tainted a little from the survey that we sent out to because most of the people bowl that we talked to in general are interested in AI in some form or another exact data data collection database always matters but it's but it is interesting because we have people who are interested in the subject but they're not necessarily able to convince their organization so even though you may be like a data scientist within a very large arts company. It doesn't necessarily mean that your company has plans to do anything..

AI Ai scientist Venice Kathleen Mulch US Ronald North America Australia Alexa Eliana Asia Europe twenty twenty Melissa Britt Africa
Global AI Adoption Trends

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

08:31 min | 2 years ago

Global AI Adoption Trends

"Hello and welcome to the AI today. PODCAST I'm your host Kathleen Mulch. I'm your host Ronald smells so one of the things we do at cognitive let occurs we produce research on the markets and landscape. We look across all the people that we've spent time talking to the use cases. The case studies all Venice. Did you spend time talking to and we ask ourselves the question once a year. Well how is he. I being adopted worldwide. What's happening in the world with people actually implementing an yeah right right are people or companies are certain regions more heavily adopting ai than others? What's really going on so a few weeks ago? We published the report called Global. Ai Adoption trends and forecast for twenty twenty. So we'll linked to it in the show notes. This one is a free download so we encourage all of our listeners to download it and check out the findings but we're going to spend some time today going through the report itself and highlight some of the key findings and maybe some unique or interesting. MM findings that we found from the survey and the report that we did and so the way that we accomplished this as we serve it we sent out the survey to over fifteen hundred individuals individuals various different companies and countries all over the world and got some two hundred and something responses and use those responses to one inform US specifically what's happening of course allowed allowed to also generalize in some ways about the trends that we're seeing for that if you're interested by the way participating in future surveys. I encourage you to reach out to us. Send us an email to info so I- NFO at Melissa and L. Y.. Dot Com and we would be happy. Include you in future service right so so we'll go through some of the key findings first and then the way that we also broke. This down was by our seven patterns obey because we said okay. Well let's one thing to say people are adopting ai but how truly are. Are you adopting A. Are you doing predictive analytics. Application or chat BOT application or hyper personalization application so some of the key findings that we've found from the the report is that by twenty twenty five over forty percent of the respondents that answered our survey said that they will implement a I in one or more of the identified seven patterns earns of Ai and almost ninety percent said that they'll have some sort of impress a implementation over the next two years. We found those numbers to be. You know very positive positive signs for the industry because it saying almost half will implement one or more pattern by two thousand twenty five so just a few years and then ninety percent so nine out of ten said that they'll have some sort of a m progress within the next two years so you might be thinking. He's looks like contradictory information. Wise that the forty percent or say and do but ninety percent have a project that they're doing two years you would think that ninety percent they're doing well this has to do with of course understanding versus the patterns because we asked them all. Okay yeah great well. Let's not talk about. How many of you are looking at doing a chat Bot and the next two years okay? How many of you are looking at doing a machine learning predictive analytics or recognition project or some sort of automation project? That's using cognitive automation. Oh well all of a sudden now. The numbers started going up and in our chart will show you kind of how the adoption patterns are looking looking because basically when you start looking at the more details yeah machine learning honestly an ai are being embedded in everything and it actually may be difficult to avoid using a machine learning. So even if you're saying well maybe we're not intending to build her on machine learning models it may end up being the end up using them anyways so we also sort of looked looked at turtle how the world was moving with. We're like well. Maybe you know North America Europe bird kind of moving at a different pace Rasiah Africa. Now you know what this is one of the interesting things about I in our research from all respondents and a response come from all over the planet they are all roughly moving at the same pace. It is true that you know Australia. Eliana Asia Europe. They have different timing. What their plans? But basically it's not like we're seeing an over concentration of aggressive plans and North American Europe and less so otherwise it's just it seems like this is just the global movement and then another thing we've talked about process automation a a lot and many companies especially many government agencies here in the. US are very hot and heavy on and process automation. In general. What we found is that fifty four percent of respondents plan to implement a approaches to process automation within the next few years so over half and then fifty two percent of respondents plan into implement a enabled conversational systems by twenty twenty five so those again were not really surprising numbers for us but something that we wanted to point out because because people are finding value in automation and I think in general taking their data cleansing it and then using it for higher level value uh-huh and so when they're able to take cognitive approaches in process automation they're starting to really see value in so we're excited that people who are really moving forward with that? Let me talk about a enabled conversational systems all the time and how companies can use that to help in a variety of different ways that can help with customer service service can also help with. It self service so they could use it internally as well and it's able to allow companies to do more with the same or less resources than before right right so our last sort of key finding sort of digging so some of the more details in a moment is that for the organizations that are sort of struggling with making ai ham or like a haven't quite quite yet taken the step. What they've said is that their biggest barrier to adoption is actually insufficient quantity or quality of data? That's like one of the biggest things things followed by lack of talent so basically people in data for a lot of response to send you know even for the people who are moving ahead. They have acknowledged that these are things that a slowing them down so for the companies that are not planning to implement ai at all within the for the next two years. The thing that they said was the biggest showstopper was just. They haven't yet justify the Arwa. which kind of makes sense or that? There isn't enough of an advantage of AI. Over non approach sprang for the ones that have taken that next episode okay. I think there's I want to do this project you know. I have have an R. Y.. I think it's going to give me an advantage. There getting stuck on people and data right right and that really comes as no surprise because cleaning data it can be a very very manual process very time intensive it can also be very costly as well and depending on the sensitivity of that data. That depends on what vendor or you can go with and where physically the data needs to be cleansed and crapped then followed by limited availability for a talent and skills. We've talked about. There's a big telling cruncher especially around data scientists so some of these smaller organizations. Just don't have the money to afford a data scientist on their team. So what can and they do then. That's where they're limited by hiring talent so digging a little bit deeper in one of the things we did our report is we asked them say okay. Well how many of you doing okay. Great Right now. How many of you are doing hyper civilization or pattern about is a nominally? Is predictive analytics. No automation which is not AI. We spent many reports talking about that. But we do track it because it is that when those pathways to get Aso we talk about process automation separate from autonomous systems. And then we talk about conversational systems recognition recognition systems and then goal driven system. And maybe it might not come as much of a surprise but the thing that's been the most widely implemented as of last year conversational national systems chat bots voice assistance Alexa skills. You know smart tech spots and embed. Yeah because you may not necessarily be thinking of those things is they. They're are all powered by machine learning especially the constitutional system. I you know high rate of adoption moving at a very sort of steady in four grants annual growth that the annual growth but like the OT overall adoption is like twenty percent and twenty four percent just keeps growing that episode. The thing that's kind of more interesting is the ramp. The rate at which people are implementing running process automation right or plan to within the next few years so in twenty nineteen there were about. Ten percent of participants had are bought in production production. Twenty twenty about sixteen percent but by twenty twenty five so five years from now fifty four percent so one out of every two every other. Yeah half half of the people want to have. Rpa In practice and implementation at their various companies and that says a lot to the growth in the potential of that

AI Venice Kathleen Mulch United States Ronald Twenty Twenty North America Australia Eliana Asia Europe Melissa Alexa Scientist Africa
Benchmarking the Voice Assistants

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:14 min | 2 years ago

Benchmarking the Voice Assistants

"Hello and welcome to the AI today podcast. I'm your host Kathleen. Mulch and I'm your host. Ronald smells are so on today's Today's podcast. We're going to talk about a report that we've recently released. That's actually the second version of a report that we released last year. which is an update to our voice assistant benchmark mark and you know we always thought it was funny? You Talk to these voice. Assistant like Amazon selects are apple or Microsoft Cortana or Google Home Google assistant device vice and sometimes they give remarkably good answers and good results and sometimes they just don't right right and sometimes they give out interesting results as well. You're like why would you give this as an answer sometimes if you ask black color of the rainbow or something and talks about black how my soul is going off on a tangent like go out that my child thinks it's funny but what we realized that we know that the technology that powers voice assistance. There's actually two parts right. There's the part that like tries to understand what you're saying like converts the Audio Way Form into words right or at least tries to and then at some point in trying to understand well. What do those words mean and tries to? I understand what you're intending and then generate some sort of response back and they're actually two different things right. There is the text to speech speech to text and natural language processing and there's the whole natural language understanding and we're we're like well. That's the part we care about the most because we know that these devices are getting better at understanding what we're saying right right and so we actually decided that we weren't going to test it on its ability we to understand what the human was saying so to take away any ambiguity with this benchmark and to make sure everything was fair. We ended up using a voice generator raider and we did that with last year's benchmark as well. We think it worked really well so we did it again so just to back up a little bit here. What we're talking about is that last year we put together a benchmark where we ask in a series of eleven sets of questions roughly ten questions each we ask these voice assistance using voice generator to avoid accents and variable pitch or whatever we ask a series of questions and a bunch of different categories and what we try to ascertain here rather than the quality of the natural language processing were trying to basically figure out are these systems systems actually intelligent assuming that like maybe even if we just typed in the question so that we take all of the variability of didn't understand the way for him even if you did that can the system actually understand what you're trying to ask and provide an answer and that's what we wanted to test was the intelligence behind the voice assistance. which actually aren't the devices themselves at all? It's actually the cloud based systems that power these devices and the other things that are from these organizations right so this year we expanded the benchmark and Mark Little and we asked twelve sets of questions for one hundred twenty questions towed Jordan forty four because we had a couple of part important the roughly about one hundred red and twenty two hundred and in that we tested four voice assistance which we found are the most popular that's why we tested them so it's Amazon Alexa Google Assistant Google Home Apple Siri and Microsoft's Cortana and we wanted to ask the question and get the answer to just how intelligent is as the AI back end and as mentioned yeah. That's really what we're testing and so these questions really get to. We'll talk about the different kinds of questions that we're asking and what we were trying to get on them. It's not like we really wanted to know. How many doughnuts are there in a box of a dozen donuts. I think it's Sorta like that's something you ask three year old. Hopefully give you an answer. That might be a little hard carver three. Maybe but the idea is like what you really want to know is. Can these systems take apart the question understand first of all. What is it that you're asking about? Okay how many okay so it's asking awesome for quantity and then you'd have to listen to the rest of the sentence and realize that you're it's something in the sentence you're asking for and it turns out. That's actually kind of a hard thing to do. Because first of all you can't program program that because of the literally infinite number of ways that people can combine words together so has to be machine learning there has to machine learning answer to this right. That's for those follower our podcast machine learning as a way of teaching computers to do through examples and data and not through programming right the fastest definition. You'll hear anywhere but we're actually really testing those machine learning earning models. That's what we're really trying to figure out right right and so that's why we wanted to make sure that the voice didn't matter and so we were testing

Google AI Apple Microsoft Amazon Kathleen Ronald Mark Little Jordan Three Year
"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

04:31 min | 3 years ago

"kathleen mulch" Discussed on AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

"The I. Today podcast produced by cowed militia cuts through the hype and noise to identify what is really happening now in the world of artificial intelligence learn about emerging trends technologies and use cases from Cog Melinda Ika analysts and guest experts hello and welcome to the A._I.. Today podcast. I'm your host Kathleen Mulch and I'm your host Schmeltzer. We're here live at the air world all government conference in Washington D._C.. It's June twenty-fourth to twenty six two thousand nineteen and we are thrilled to have our guest today aren't the guard who is the emerging market and technology director at cray thrilled to have here. Welcome to the PODCAST. Thanks it's great to be here. Yeah welcome arthy and thanks for joining us. We'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background and your current role at cry. Okay I joined cray a little bit over a year ago to head the development of our market and technologies technologies lash products strategy in A._I.. I'm part of a sort of unusual team at cray headed by the V._p.. Of A._I.. And I say unusual because we're very much focused on us specific application space to identify how we can move into this application segment. My role here compared to my background is a bit of a change for me until as recently as a year ago I was working primarily as data scientist and in data science team lead roles and so it's a bit of a transition and to be more in the strategy space on the business side of things however I will say that as a data scientist I frequently found that maybe a quarter of the role is actually thinking about product and actually thinking about business because you really need to think about how people are using data science and what they do so in that sense. It's not a major change for me prior to working as a data scientists in the commercial sector. It's very exciting for me to be here at A._I.. World Government Matt because I used to be a civil servants. I used to work at the Office of Management and budget overseeing our efforts at the Department of Energy and advanced computing so this is really exciting remmy. You got to know that you also have a rich background. Maybe can go back because you have a number of degrees and they just for the sake of our listeners are you can tell us a little bit about that passed and how to brought you to kind of where you are right now yeah so my background is an astrophysics and in aerospace engineering. I have a P._H._d.. In astrophysics and sometimes I- jokes that I was a data scientists before for that was a term because as a researcher and I won't talk about exactly when but I was dealing with fairly large data sets at the time maybe today they wouldn't seem super large but at the time multi terabyte did us that's where very very large and we were developing the techniques techniques to process that data very quickly as it came off of telescopes and then developing models to make decisions about what to do at the telescope the next night so it's been fun and at a world governor. I know that you gave a talk here so for our listeners that weren't able to join us. Can you give quick to minute recap of your talk definitely so I was speaking and part of a seminar on how agencies or even commercial entities might choose which vendors to use the A._i.. Space Because A._I.. Vendor landscape is fairly complex and the purpose and the focus of mytalk was really to say actually don't start thinking about this overwhelming landscape of three thousand plus vendors start by really defining what it as you want to do and once you've defined what it is you want to do you can start to think about what the data you have would help solve that problem and once you know what you're trying to do and what some of your data needs are. It becomes much easier to select the right technology. Cheap under your that's interesting because technology landscape you write about this quite a bit cod liver. We track about three thousand vendors in the landscape which is actually a subset of this is just an A._i.. Machine Learning Space and that like seventy percent are like these industry and domain specific things applications occasions to specific like medicine and sales and marketing and finance and cyber security and agriculture right and then of course there's enabling technologies. How are you seeing the sort of landscape of A._I.? Vendors and kind of the diversity of ways in which is being applied. That's a great question. I think you almost already hit on it. In the sense that there is kind of almost bifurcation in the vendor landscape where I tend to think of there being platform solution providers and so they may be infrastructure platform providers or they may be software platform form providers <unk> generally enable people to Adopt A._I...

scientist cray Melinda Ika Kathleen Mulch Washington Schmeltzer technology director Matt Office of Management Department of Energy researcher seventy percent