Thanks! Share it with your friends!

MIT 6.034 Artificial Intelligence, Fall 2010
View the complete course: http://ocw.mit.edu/6-034F10
Instructor: Patrick Winston

Can multiple weak classifiers be used to make a strong one? We examine the boosting algorithm, which adjusts the weight of each classifier, and work through the math. We end with how boosting doesn’t seem to overfit, and mention some applications.

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

Reed Jessen says:

This is such a clear path to understanding. Thank you, Prof.Winston.

Hélder Sousa says:

Right aisle : 2:38 He's exited.

Jason Shi says:

Handwriting model

Myren Eario says:

At 16:25 doesn´t the orange line at the bottom symbolize the exact same thing as the orange line at the very left? Both say "Everything is +" or "Everything is -". And then we don´t have 12 classifiers but only 10.

Raphael Seitz says:

Why is a coin flip a weak classifier if p1>p2 with p1+p2=1? 0.5×p1+ 0.5×p2 still is 0.5.

sandesh jain says:

I didn't get the part where new weights are scaled to 1/2 what good does it do ?

emanuelen5 says:

A comment on the transcription: A lot of times when it is transcribed [Inaudible], he says "Schapire", which is the inventor of Boosted learning (Robert Schapire)

Anurag Sodhi says:

Thanks for amazing lecture!

אמיר דוידוף says:

an amazing lecture ive enjoyed every second.
question: would this work well for classification with very unbalanced data set?
minority class at about 1 percent

Why is there a sheep on the first row?

JPatrick Davenport says:

How do the stumps tighten back in?

iliTheFallen says:

Perfect teaching! Great job, Sir.

Moustafa Banbouk says:

Way to go Doctor, the explanation is very clear and unique. I was just wandering if anyone has an idea what application was being used to demonstrate the algorithm.

Niels W says:

The not overfitting thing is really mind blowing, because it seems to me like the VC dimensionality of the demonstarted classifier is infinity. I was about to write a question like this:
Does the volume of the space of which the classification result depend on an outlier decrease in any case, or are there cases (of low probability) in which they occupy more volume?
I guess that the volume decreases, if there are good samples around the outlier, and that the volume can stay large if the outlier lies far away from the subspace in that the good samples lie. If that holds, it is still unlikely to get test data points in that volume even if it stays large.

Evan Kozliner says:

It's incredible that there are even empty seats in this lecture. Truly an amazing professor

yesnoyesnookay says:

What do you mean by "data exaggeration"?

gumikebbap says:

so how does the program choose the number of classifiers to use?

Tim Dries says:

Man, that straight line on the board in the beginning, what a pro

just awesome

ali fawzi says:

i would like to thank you about your fantastic contribution in the all science &especially in computer field

Eduard Barnoviciu says:

Oh, wish I'd learn this in college. Close but not quite. Thanks MIT I guess

GigaFro says:

Phenomenal lecture. Easy to understand and as said before, great hand writing. Thanks for sharing, it is much appreciated 🙂

q zorn says:

so to solve a data set, is there a program to 1st determine which neuro software is the correct choice to produce the correct results, KNN, SVM, Boosting, etc…?

apanapane says:

Thank you for this lecture.

Vikram jain says:

Last part of Thank God Hole is excellent explanation

Yuchen Li says:

switch speed to 1.25 😀

Omid Mo says:

An outstanding teacher. I appreciate Dr Winston. He explains confusing stuff in a very simple way.

Neda hajiakhoond says:

Great teacher!

DARPA SUPERHIT 2021 Play Now!Close

DARPA SUPERHIT 2021

(StoneBridge Mix)

×