(upbeat electronic music) >> Narrator: From the Fairmont Hotel in the heart of Silicon Valley, it's theCUBE. Covering when IoT met AI, The Intelligence of Things. Brought to you by Western Digital. >> Welcome back here everybody, Jeff Frick here with theCUBE. We are at downtown San Jose at the Fairmont Hotel. When IoT met AI it happened right here, you saw it first. The Intelligence of Things, a really interesting event put on by readwrite and Western Digital and we are really excited to welcome back a many time CUBE alumni and always a fan favorite, she's Janet George. She's Fellow & Chief Data Officer of Western Digital. Janet, great to see you. >> Thank you, thank you. >> So, as I asked you when you sat down, you're always working on cool things. You're always kind of at the cutting edge. So, what have you been playing with lately? >> Lately I have been working on neural networks and TensorFlow. So really trying to study and understand the behaviors and patterns of neural networks, how they work and then unleashing our data at it. So trying to figure out how it's training through our data, how many nets there are, and then trying to figure out what results it's coming with. What are the predictions? Looking at how the predictions are, whether the predictions are accurate or less accurate and then validating the predictions to make it more accurate, and so on and so forth. >> So it's interesting. It's a different tool, so you're learning the tool itself. >> Yes. >> And you're learning the underlying technology behind the tool. >> Yes. >> And then testing it actually against some of the other tools that you guys have, I mean obviously you guys have been doing- >> That's right. >> Mean time between failure analysis for a long long time. >> That's right, that's right. >> So, first off, kind of experience with the tool, how is it different? >> So with machine learning, fundamentally we have to go into feature extraction. So you have to figure out all the features and then you use the features for predictions. With neural networks you can throw all the raw data at it. It's in fact data-agnostic. So you don't have to spend enormous amounts of time trying to detect the features. Like for example, If you throw hundreds of cat images at the neural network, the neural network will figure out image features of the cat; the nose, the eyes, the ears and so on and so forth. And once it trains itself through a series of iterations, you can throw a lot of deranged cats at the neural network and it's still going to figure out what the features of a real cat is. >> Right. >> And it will predict the cat correctly. >> Right. So then, how does that apply to, you know, the more specific use case in terms of your failure analysis? >> Yeah. So we have failures and we have multiple failures. Some failures through through the human eye, it's very obvious, right? But humans get tired, and over a period of time we can't endure looking at hundreds and millions of failures, right? And some failures are interconnected. So there is a relationship between these failure patterns or there is a correlation between two failures, right? It could be an edge failure. It could a radial failure, eye pattern type failure. It could be a radial failure. So these failures, for us as humans, we can't escape. >> Right. >> And we used to be able to take these failures and train them at scale and then predict. Now with neural networks, we don't have to take and do all that. We don't have to extract these labels and try to show them what these failures look like. Training is almost like throwing a lot of data at the neural networks. >> So it almost sounds like kind of the promise of the data lake if you will. >> Yes. >> If you have heard about, from the Hadoop Summit- >> Yes, yes, yes. >> For ever and ever and ever. Right? You dump it all in and insights will flow. But we found, often, that that's not true. You need hypothesis. >> Yes, yes. >> You need to structure and get it going. But what you're describing though, sounds much more along kind of that vision. >> Yes, very much so. Now, the only caveat is you need some labels, right? If there is no label on the failure data, it's very difficult for the neural networks to figure out what the failure is. >> Jeff: Right. >> So you have to give it some labels to understand what patterns it should learn. >> Right. >> Right, and that is where the domain experts come in. So we train it with labeled data. So if you are training with a cat, you know the features of a cat, right? In the industrial world, cat is really what's in the heads of people. The domain knowledge is not so authoritative. Like the sky or the animals or the cat. >> Jeff: Right. >> The domain knowledge is much more embedded in the brains of the people who are working. And so we have to extract that domain knowledge into labels. And then you're able to scale the domain. >> Jeff: Right. >> Through the neural network. >> So okay so then how does it then compare with the other tools that you've used in the past? In terms of, obviously the process is very different, but in terms of just pure performance? What are you finding? >> So we are finding very good performance and actually we are finding very good accuracy. Right? So once it's trained, and it's doing very well on the failure patterns, it's getting it right 90% of the time, right? >> Really? >> Yes, but in a machine learning program, what happens is sometimes the model is over-fitted or it's under-fitted or there is bias in the model and you got to remove the bias in the model or you got to figure out, well, is the model false-positive or false-negative? You got to optimize for something, right? >> Right, right. >> Because we are really dealing with mathematical approximation, we are not dealing with preciseness, we are not dealing with exactness. >> Right, right. >> In neural networks, actually, it's pretty good, because it's actually always dealing with accuracy. It's not dealing with precision, right? So it's accurate most of the time. >> Interesting, because that's often what's common about the kind of difference between computer science and statistics, right? >> Yes. >> Computers is binary. Statistics always has a kind of a confidence interval. But what you're describing, it sounds like the confidence is tightening up to such a degree that it's almost reaching binary. >> Yeah, yeah, exactly. And see, brute force is good when your traditional computing programing paradigm is very brute force type paradigm, right? The traditional paradigm is very good when the problems are simpler. But when the problems are of scale, like you're talking 70 petabytes of data or you're talking 70 billion roles, right? Find all these patterns in that, right? >> Jeff: Right. >> I mean you just, the scale at which that operates and at the scale at which traditional machine learning even works is quite different from how neural networks work. >> Jeff: Okay. >> Right? Traditional machine learning you still have to do some feature extraction. You still have to say "Oh I can't." Otherwise you are going to have dimensionality issues, right? It's too broad to get the prediction anywhere close. >> Right. >> Right? And so you want to reduce the dimensionality to get a better prediction. But here you don't have to worry about dimensionality. You just have to make sure the labels are right. >> Right, right. So as you dig deeper into this tool and expose all these new capabilities, what do you look forward to? What can you do that you couldn't do before? >> It's interesting because it's grossly underestimating the human brain, right? The human brain is supremely powerful in all aspects, right? And there is a great deal of difficulty in trying to code the human brain, right? But with neural networks and because of the various propagation layers and the ability to move through these networks we are coming closer and closer, right? So one example: When you think about driving, recently, Google driverless car got into an accident, right? And where it got into an accident was the driverless car was merging into a lane and there was a bus and it collided with the bus. So where did A.I. go wrong? Now if you train an A.I., birds can fly, and then you say penguin is a bird, it is going to assume penguin can fly. >> Jeff: Right, right. >> We as humans know penguin is a bird but it can't fly like other birds, right? >> Jeff: Right. >> It's that anomaly thing, right? Naturally when are driving and a bus shows up, even if it's yield, the bus goes. >> Jeff: Right, right. >> We yield to the bus because it's bigger and we know that. >> A.I. doesn't know that. It was taught that yield is yield. >> Right, right. >> So it collided with the bus. But the beauty is now large fleets of cars can learn very quickly based on what it just got from that one car. >> Right, right. >> So now there are pros and cons. So think about you driving down Highway 85 and there is a collision, it's Sunday morning, you don't know about the collision. You're coming down on the hill, right? Blind corner and boom that's how these crashes happen and so many people died, right? If you were driving a driverless car, you would have knowledge from the fleet and from everywhere else. >> Right. >> So you know ahead of time. We don't talk to each other when we are in cars. We don't have universal knowledge, right? >> Car-to-car communication. >> Car-to-car communications and A.I. has that so directly it can save accidents. It can save people from dying, right? But people still feel, it's a psychology thing, people still feel very unsafe in a driverless car, right? So we have to get over- >> Well they will get over that. They feel plenty safe in a driverless airplane, right? >> That's right. Or in a driveless light rail. >> Jeff: Right. >> Or, you know, when somebody else is driving they're fine with the driver who's driving. You just sit in the driver's car. >> But there's that one pesky autonomous car problem, when the pedestrian won't go. >> Yeah. >> And the car is stopped it's like a friendly battle-lock. >> That's right, that's right. >> Well good stuff Janet and always great to see you. I'm sure we will see you very shortly 'cause you are at all the great big data conferences. >> Thank you. >> Thanks for taking a few minutes out of your day. >> Thank you. >> Alright she is Janet George, she is the smartest lady at Western Digital, perhaps in Silicon Valley. We're not sure but we feel pretty confident. I am Jeff Frick and you're watching theCUBE from When IoT meets AI: The Intelligence of Things. We will be right back after this short break. Thanks for watching. (upbeat electronic music)