What's in Your Wallet? For Capital One, the Answer Is AI - Ep. 70


The. Hello and welcome to the Invidia a podcast. I'm your host, Noah Kravitz what's in your wallet? Does that tagline ring a bell? I might be dating myself here, but when I hear the words Capital One, I immediately picture Jennifer garner or maybe Samuel l. Jackson pitching credit cards and TV ads asking what's in my wallet. Well, it's easy to associate Capital One with credit cards and other consumer services. There's nothing wrong with that. The banking giant is also doing a ton of work with machine learning and AI on fronts, ranging from fraud detection to forecasting to customer service here to talk to us about how AI is impacting the finance industry is Nissan mckell, managing vice president of machine learning at Capital One. Nissan, thanks for joining the Invidia podcast. Thank you. Glad to be here. Let's start kind of broadly capital one's a Bank finance company. Why does the Bank need AI and machine learning? What are you doing? Kind of day to day and more. Broadly at Capital One and how our AI and machine learning lane important role in your work yet? It's a good question. It's funny to think of a Bank to being a and machine learning. I admit Capital One is the first Bank working at. I came from a healthcare and did machine learning there for a long time. So my first thought when capital want to broach knee was, will it make sense lease machine learning to predict lending decisions and other financial forecasting, and they'll be about it. So I wasn't terribly enthusiastic only because it seemed pretty specific and pretty myopic with Billy turned me onto the opportunity. What let's make cited every day now is the fact that we use a and machine learning pretty much every facet of our business may be customer facing applications like fraud monitoring automation of internal. Processes like our call center up rations and there's a ton of work redoing their customer experience, digital marketing, basically anything you could think of even image processing and computer vision. It all comes to play in our application of machine learning. So can you give us an example without, you know, without disclosing any trade secrets, so to speak. But an example of one of these applications in concrete use. Sure. So a big focus we have is leading the next era Banking's through breakthrough digital experiences, and we want those experiences to operate Bill time, and in order to do that, you really have to have them powered by machine learning, no way of making that happen otherwise. So from a customer standpoint were leveraging machine learning to help customers become more financially empowered. We want them to be able to make predictions about upcoming bills, deter. Expenses that may reflect something that isn't consistent with what they would normally want to do, or that's even signal of fraud, a better manage their spending. And of course enhance their digital experience and their satisfaction with us as a consumer. Are any of these applications available now to your customers? Yeah, we have, for example, e-e-e-e-no is our chat bot we were the first one to roll that out in a major Bank and that obviously delivers recessional. I delivered some of the forward leaning promise that machine learning has been able to deliver on with deep learning. So it sounds like you're covering a lot of bases from business side predictions, like you mentioned fraud detection obviously is a huge problem for consumers and banks like lending decisions, those sorts of things, and then getting more into the customer experience, you're covering kind of a lot of ground. How does the work that you're doing at the company. Speak more broadly to, you know, current trends and the fields of AI and machine learning? Yes. So if I think about fraud as an example, that is what we really at the center of the pattern. We use a lot of the applications of machine learning guy. Currently going on as security is really tends to fi as anti money laundering. Efforts of had intensified anomaly detection from systems perspective is really kicked up in the last couple of years industrywide. I would say that banking is rapidly catching up with some of the technology giants in terms of how we apply machine learning in fraud related areas more broadly speaking. I think that some of the trends we're observing in terms of GP you use in terms of taking. Deep learning from what was relatively niece a couple of years ago to pretty much mainstream across the board. We started out with deep learning only in our process processes space have now expanded far beyond that really were following a the overall mythological trends that are probably true for most industries of figuring out the best way to create value with the s. civic assets that we have. So as you mentioned, there's been this rapid uptick actively and interest in AI and deep learning across multiple industries. And you know, as you said before, getting into banking and finance, you came from the health industry, but we've had several guests on the show over the past few months who've really pointed to the past, oh, say, eighteen months to two years as being this really fertile period of kind of an accelerated growth pace and and they've pointing to hardware advances. As being one of if not the key driver. And I'm just wondering given a special that you've kind of span two really big industries and what you were just saying about uptick of use and interest in a deep learning and banking. Have you seen the same recent growth in a I broadly, and then would you also point to hardware and hardware advances as being a key driver or if not, you know what else have you seen? Absolutely. I think from my perspective, both in healthcare and financial services, what has been pushing in this direction for the past couple of years is really the need to deliver wheel time experiences. People are now pretty much accustomed to everything being Bill time, and that's what they wanna see and the kind of distributed computing infrastructures that we had previously just couldn't cut it. So if I think of our fraud prevention experiences, in example, we currently. Are delivering real time fraud prevention at chore customers starting all the way from interactive alerts so customers can feel protected when they're learned to potential fraud. They can easily report fraud transactions. They can lock their card in real time on the spot. They can continue spending on ballot birches again in real time and receive verification. A new card is on the way. So we're leveraging structured, but also massive, unstructured data to detect one transaction is potentially fraudulent, and we've already been able to reduce false positive rates dramatically. You can imagine that doing all that in real time just requires a different kind of hardware than was previously needed. Sure. Apart from the day to day application, like you mentioned, consumer experience and fraud protection, obviously a perfect day to day up occasions of a. Are you doing any research in the field. Capital One? Yeah. So obviously there's not gonna be a ton of innovation. If there isn't a lot of research, there's probably a lot more research that ends up not going into strict production because there's learning that doesn't generate a product right away. Of course, you're probably thinking of some of the investments that are farther out and that we definitely have number of investments going on. For example. The explainable AI effort at the forefront where we're trying to innovate. You can imagine as a Bank where fair lending is incredibly important to us, eliminating bias is incredibly important to us a figuring out ways in which we can deliver machine learning in a way that is unconstrained by pay city. The way a lot of the deep learning, for example, approaches would traditionally be constrained. Something were invested in our guest today is Nitzan Mikhail. He is the managing vice president of machine learning at Capital One after spending chunk of his career working with AI and deep learning and data in the health industry. He's made the shift over to banking services. So you've been with Capital One for roughly a year now. All right. Yeah, let's about right. So if we go way back, how did you get into a field of AI. Admission learning? Yeah. So I'm gonna date myself the way you did. I started in this field before it was called machine learning data science at the time. It was just around the time of the human genome project, and it was one of the early applications of high performance computing to massive data sets that were growing really badly. And there were this realization that the way we were doing things previously just not going to cut it had to develop neutrals in new approaches altogether. That's how I got into it. I worked for quite some time in the genomic space and then moved on into medical technology because streaming data bears really literally a matter of life and death. And so some of the decision you can make with machine learning just can have such a huge impact. So you mentioned that. You spent some time around the time of the human genome project working in that space with genomes, were you involved in the human genome project itself or if not, what were you doing? And is there kind of sounds like a fascinating area of research obviously impact, but then also kind of with the growth of high performance computing. Is there a story is kind of a sixty second version of what you were doing that and maybe some lessons learned Schorr again, dating myself. It's going to sound babe, primitive. Now I was on the human genome project itself. We I was doing comparative genomics. That's where you're comparing long sequences of DNA between individuals. A lot of this has to do with disease detection and genetic mapping, and then even between species which is some of the work I did as well. What you're trying to do is identify anomalous patterns as well as association patterns. At between long stretches of DNA in some both marked and unmarked labels. Even though we had access to relatively as sophisticated hardware, it was something where we would send it off for analysis or start the analysis than have to wait for two weeks, some Celts back, right? So honestly, I even played around with neural nets at the time. It wasn't very promising, but it was just conceptually such a cool idea. Really. There just wasn't the kind of computing power that would allow you to get very far there as the most of it was random, fours decision trees in even that would take days if not longer to get any results back. If you're in that head space now, then sort of fast forwarding to two thousand eighteen was kind of a period or even a moment where you sort of saw not the late at the end of the tunnel, but almost like a breakthrough to a new tunnel with light at the end of it because of sewing. The happened in the hardware world or suddenly these two week waiting period. You kind of realize like, oh, this thing is going to change all of that. Yeah, I wish it was a single lightbulb moment. There isn't quite that, but I would say when really the move to the cloud became inevitability that's when things changed. Now I was in a Billy regulated environments still him. So things were slow. I think we're probably somewhere hind her, but probably around two thousand ten or so maybe two thousand eleven is when I started seeing that light at the end of the tunnel and could really conceive of using some more sophisticated methods with the kind of data we working with. And since thinking about the present day and not so much the applications of what you're doing, but purely get your head on purely from technical perspective, what kinds of challenges are you grab. With there's something that you know gets you up in the morning that you really psyched to kinda chip away at, and and you know whether it's something you're architect ING or a more specific problem you're solving or anything that kinda gets your geek crank going as our producer like, say. Yeah, I actually really like thinking about how problems that there is business stakeholders think is really unique and really different is actually not that unique. It all I know that sounds down or. I find it really cool seeing patterns across problems that indicates a common solution. Right? Because the last thing we wanna do is be building one off solutions every time. And so looking at anomaly detection in the fraud space anomaly detection in the customer experience, fates seeing that that pattern is the same and we can go after it in the same with the same solution is is is what gets me up and gets me going. So taking from their now looking ahead, let's say the next five years, what do you think some of the forward-looking implications of your work or even sort of a in the finance and financial service industry? Kind of more generally, you know, what might the implications be? Where do you think this is headed in the near term? Yeah, I think the jury's subtle out in many ways, but its most basic form, I think machine learning is doing now. It's collapsing away the distance. Between the companies and their consumers. So in our case, the Bank and our customers, and with that distance going away where able to really understand the customer and react to what the customer needs in a very immediate, what way? And so and course correct in a really immediate way as well. And so what that means is the whole relationship between a customer and the Bank changes, because now we can be a real advocate for them. The most direct possible way while before there was just this big gap in customer was for lack of a better word was sort of a number. Sure. Now we can really understand their context and the machine learning aspects of that, because I've heard that similar sentiment expressed by companies that do CRM software other platforms from your perspective is interested there so many customers and so much data in the financial space that you really need that high-performance machine learning to be able to make sense of it, or how does the NFL play into getting closer to the customer? Yeah. No, I would say it has much more to do with instrumentation, so actually being able to see what the customer is needing in a much more ongoing basis than it's also being able to build machine learning models that take that full account to the. Customers history with us. So we get to know them better better over time as some of your listeners will know, for example, building a large scale production grade time series anomaly detection platform is really hard cases. It's incredibly intensive, but it also scales really unfavourably like the longer the history of a customer or any data set the longer this sequence, the harder that problem gets and it scales very babe released. So now with some of the hardware improvements, whether it's cheap us or even some of the older distributive computing solutions, we're able to do that in much greater scale or at much greater scale and therefore big much closer to the customer. So it's less about nothing against the Iran, but I feel like that's very two thousand eight. So apart from the the day to day applications of AI and machine learning that you just spoke to, I know that research development are Indies part of, you know any any big technological effort. I'm sure Capital One has machine learning research going on the time I can you speak to anything that you're doing that isn't necessarily gonna show in a product, you know tomorrow or maybe never as research goes, but things that you're doing with an eye towards the long haul. Sure. So brought the Capital One has machine learning at the very core of its are India Genda. Some of the key areas working on our accountability, explain ability things like explainable, AI, ethics, fairness governance, of course, security and privacy are incredibly important as well. We partner with some of the top universities to further these research areas in critical elements of our business and also the machine learning. Theory that would help advance our ability to interact with regulators and explain what our agendas I can go into a little bit more on the explain ability. In fairness, if if you think that would be interesting to listeners, I actually was going to ask wanna trip over my own tongue and ask you to explain I explain ability, but yeah, that'd be great. Sure. So being a heavily regulated environment, we wanna make sure that we're not just meeting the regulatory requirements, but that we helped set the standard for what a fair, an ethical machine learning deployment looks like civically in the financial services. So there's great potential to leverage data analysis the she learning to gauge credit worthiness, but we want to make sure that were balancing the use of this technology. Some of which is moral Paik than than earlier modeling efforts. If you think of learning the right. With the appropriate development application of adjacent algorithm technology that ensures fair and unbiased outcomes. So we really want to maintain the highest standards for explain ability in an ethical and fair way. As we develop more advanced models for more use cases, we wanna push the envelope when it comes the sophistication of our models and our algorithms. But at the same time, as you push that envelope, it becomes harder and harder to explain the underlying reasons, food decisions. And so we're investing a lot in be searched that will help us make sure for ourselves that were always doing the right thing also communicate to our customers in our regulators that we are doing the right thing for customers at all time, we do have a Capital One position paper on fair Linden at the effect trick workshop on responsible recommendations in October. When. When it comes to things like lending decisions, say, accept something easily grasp for people. How much of that decision making right now is being informed by machine learning system versus another, perhaps less opaque algorithm system or even human oversight, like head of those things together. So Capital One from its very beginning was very sophisticated in our decision. So we've always leverage data in a way that most banks weren't. So decision modeling has always been part of our decision ecosystem, but it's very much with a human component and always with oversight that insures there's, there's nothing that's gonna go awry in that decision framework. Now, as I mentioned were continuously improving what we can deliver from she learning perspective, but we're never. Outpacing our ability to do that in a way it's completely responsible for Capital One customers or not. Because everybody at this point has a stake in the banking system in one way or another. It's good to hear. As you said anyone listening can can attest to that sort of tension between the excitement of the system, getting more powerful and sophisticated. And also, you know that that feeling of even though I built the thing, do I fully understand where it's kind of going now with its decision making. Yeah. What's also exciting for me is banking hesitant. Traditionally been at the forefront of technology innovation in large part because of the regulations are important. So it's exciting that when it comes to explain ability, I think that we're incredibly well positioned to take the lead there even more so than our technology companies because it's really at the core of what we do, and it's incredibly important to us to make sure that that's always at the forefront of every technological innovation that we take. So for folks, listening who might wanna find out a little bit more about what capital one's doing in these fields. I machine learning deep learning is there somewhere online they can go. So we do have quite a bit research submissions, etcetera, at the berries conferences. In addition to that, we have the Capital One blog. That always summarizes that information as well as provides additional information that we would be excited for listeners to come out to. Excellent. It's on mckell. Thank you so much for making the time to speak with us about what you're doing. A Capital One, given your background working with data machine learning and and what high-performance competing was called. Prior to that, it's very cool to hear your excitement about having hop from. You know what on the service may seem like very different industries, Juno mix healthcare to finance, but getting chances sort of see know there are similar problems approaches, some outcomes, no matter the industry. Very cool. Her about. Thank you. Thank you so much, really enjoyed talking to.

Coming up next