[MUSIC] So, in the last several segments, we covered selected topics in experiment design, hypothesis testing, and statistical estimation. And, we motivated these topics, by considering the so called decline effect where scientific results turn out to be less reproducible over time. Or their effect size diminishes over time under repeated experiments. Okay, and we found that things like publication bias and the misapplication of statistical methods can explain a lot of this effect. Okay, and then we also talked about the Bayesian perspective, which allows you to incorporate prior knowledge easier, but opens the door to subjectivity leaking into the analysis, which is one of the reasons why it was largely rejected at the beginning of last century. But it's making a resurgence thanks to computational techniques that allow you to do sampling of complex distributions at scale. And we didn't go into too much detail there but I hope to get back to it before the end of the course. So in the next few segments I wanna talk about more of an algorithmic perspective that is more closely associated with the terms machine learning and data mining. Okay. So, what is machine learning? So, Pedro Domingo here at UW wrote a fantastic communications VACM article in 2012 that gave an overview in machine learning. And what he said was that machine learning is systems that automatically learn programs from data. So what he means here is that when you write a program by hand, you're explicitly stating how to map a set of inputs to some desired output. Right, you're giving a recipe for doing that, but the point is that giving enough examples of inputs and desired outputs, you can infer the program that maps it, as opposed to having to write it explicitly. So the joke is, why bother writing programs when you can learn them instead? Maybe another view is, teaching a computer about the world, but I prefer the first bullet, here. Okay, so it's worth maybe a few minutes or just one minute or so talking about what the difference between statics and machine learning is. It's probably not worth putting a lot of time into this or thought into this because terms are not necessarily precise and they grew up in some historical context and reflect different cultures rather than being some well defined notion that you can reason about. But, there's a paper, pretty long ago now, it was in 2001, that talked about a certain kind of difference, although they weren't specifically referring to machine learning. It captures what I believe to be the real salient difference here, okay? So, the model here on the left is that everybody involved in the space is trying to do the following which is observe that nature is able to map inputs to outputs. And we wanna be able to predict that computationally. Okay. So the statisticians will attack this problem by actually trying to model the world through some statistic process. In the paper they talk about data modeling. They talk about modeling the data. What they mean is modeling the process that gave rise to the data that you see. Okay, and so the idea is that once you have a model of the world, you can use it to sort of hypothetically draw more samples and reason about the entire population, okay. But another view that maybe's this more algorithmic view is look, you know, we recognize that nature is difficult to model and we don't really care if we have an accurate model of it. All we're trying to do is measure our performance in predicting the outputs that we've seen from given inputs. Okay. So it's not important that the model that we produce, that the program that we generate, actually reflects the real process that's going on in nature or not. We don't claim it does but what we do claim is that it does a pretty good job making predictions anyway. And so I think this is the notion that's more associated with the machine learning approach. [MUSIC]