Then we'll talk about the idea of estimation of regression coefficients.

Just like the sum of square, sum of deviation squares. So we'll square them.

And that would give us the sum of squared residuals.

So, we wanna minimize the difference between the observed scores on Y and predicted scores on Y, predicted by the model.

X and Y, we are still assuming that X and Y are both continuous, both normal and there's a linear association between them and we will move beyond those assumptions

Here we start implementing the delta debugging function.

Now if I want to take this objective function and multiply it by 10, so here my minimization problem is minimum of u of 10, u minus 5 squared plus 10.

So we think of Ci as picking the cluster centroid with the smallest squared distance to my training example Xi.

How does R, or how if we want to calculat e this by hand, how do we come up with these estimates?

look at an example. So here's a cost function J with theta.

59 squared, OK, and finally?

2x squared.

This is a space of the scan line in which I search, often called disparity, and for one location this is actually being minimized right over here.

This is an interesting one where we seem to have a linear relationship that is flatter than the linear regression indicates, but there is one outlier.

Again, there's variance in the population, sigma squared.

So let me just graph those.

We only have 1x squared terms, so let's write that down.

Look what's in the numerator, sum of cross products.

But of course minimizing squared distance, and minimizing distance that should give you the same value of Ci, but we usually put in the square there, just as the convention that people use for K means.

OK, now we can simplify.

Lynn example.

I swapped the sides.

Arthur Benjamin

AB 23 squared, OK. Audience

Andrew Goldberg Recently I've been working a lot on shortest path algorithms, especially motivated by GPS navigation applications, so basically how to get from A to B.

Here is what I mean, to give you a concrete example, suppose I had a minimization problem that you know minimize over a real number u of u minus 5 squared, plus 1, right.

So the standard deviation is one half and the variance sigma squared would therefore be the square of 0.5 would be 0.25.

So we put in one density.

First I'm gonna bring your attention to this RSS column.

These are other types of machine learning algorithms that we'll talk about

Moore's Law, with computers getting faster, actually says that our computational ambitions will naturally grow.

AB 987 squared is 974,169.

57,683 squared.

The taxi picked up two passengers.

It's going to look like that.

We talked about K means, and we talked about expectation maximization.

And as a result, sorting algorithm which runs in time proportional to n times log n is much, much faster, especially as n grows large, than a sorting algorithm with a running time that's a constant times n squared.

We said, oh, well, it needs a number of operations proportional, a constant fact or times in

This gives me the multi, the element wise squaring of

I'll assume that these events are independent.

And I think you're familiar with what that looks like.

Minus 2 and minus 2 is minus 4.

I peed twice.

Min

Min.