## 16. Learning: Support Vector Machines

MIT 6.034 Artificial Intelligence, Fall 2010

View the complete course: http://ocw.mit.edu/6-034F10

Instructor: Patrick Winston

In this lecture, we explore support vector machines in some mathematical detail. We use Lagrange multipliers to maximize the width of the street given certain constraints. If needed, we transform vectors into another space, using a kernel function.

License: Creative Commons BY-NC-SA

More information at http://ocw.mit.edu/terms

More courses at http://ocw.mit.edu

MIT 6.034 Artificial Intelligence, Fall 2010

View the complete course: http://ocw.mit.edu/6-034F10

Instructor: Patrick Winston

In this lecture, we explore support vector machines in some mathematical detail. We use Lagrange multipliers to maximize the width of the street given certain constraints. If needed, we transform vectors into another space, using a kernel function.

License: Creative Commons BY-NC-SA

More information at http://ocw.mit.edu/terms

More courses at http://ocw.mit.edu

### Buy/Stream:

StoneBridge - So Good ft Barnev

Mama India [Reloaded] (StoneBridge Mix)

*Related*

my apology but what is the convex thing in svm that makes the solution global not local? Could it be the cost function but i did not see it here.

Best explanation of SVM ever in the World/Internet.. Salute!

at 28:51 we get an equation to calculate w in term of the duals ai. But is there an equation for b?

Awesome lecture, I understand SVMs now. Thanks so much.

This course and instructor are amazing. This is out of date and not relevant any longer.

:O … As clear as water in the Caribbean

Tried many videos but This is Best Explanation of SVM .

What a good explanation! Thank you, it is really important to make things simpler than complex like other teachers do. I think I understood it right now! The only part is the transformation from a minimization problem to a max. problem, where I think more explanation is useful. But this is more opt. stuff. Thank you Prof!

Why is the professor panting so much? I hope he's healthy

could someone please explain to me why 8:40 to 9:30 is the way it is? I'm so lost on this step… why did it turn from >=0 to >= 1

Why is this dinosaur using chalk?

I liked the story he ended the lecture with. Lecturing is quite a talent.

I was watching the video and thinking "Wow, the pace of the lectures at MIT is pretty fast. These students must be really bright to follow the professor. No wonder that I'm not studying there". At the end, I found that, unbeknownst to me, I was watching it all along at the 1.25 speed.

Is that Anderson Cooper in the front row?

Someone give this man a medal. Pure brilliance. Thanks for sharing.

loved it,thank you so much

Does Winston still thinks SVMs are better then NNs / DNNs?

If Y(w*x+B)-1=0, and y = -1 => w*x = 1+B.

1-B-(1+B) = -2B

The math errors are unbearable.

dot(x,w) ?= dot(w,x)

i get lost in this part

At 11M and 33 S, Prof. Multiplies two basic equations with Y_i and forgets to attach W to X_i. Keep patience, Prof. Just forgot, student points to that one and then Prof. Corrects that. I already kept thinking for a while that oh man, what's going on here and then decided to move and see what happens!

And how do I compute 'b'?

didn't know Philip Seymour Hoffman was professor at mit…

Great chalk writing/drawing.

Good, very intuitive.