International Conference on Multimodal Interfaces

Tutorial on Statistical Machine Learning with Applications to Multimodal Processing

author: Samy Bengio, Google

Categories

Top: Computer Science: Machine Learning: Statistical Learning

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Tutorial on Statistical Machine Learning
0:01	Outline of the Tutorial
2:12	Part I
2:15	What is Machine Learning?
2:19	What is Machine Learning? (Graphical View)
2:38	What is Machine Learning?
3:45	Why Learning is Difficult?
3:50	Why Learning is Difficult?
5:02	Why Learning is Difficult? (2)
5:45	Why Learning is Difficult? (3)
6:23	Occam’s Razor’s Principle
7:58	Learning as a Search Problem
9:43	Types of Problems
9:47	Types of Problems1
10:24	Types of Problems2
10:47	Types of Problems3
11:25	Applications
11:28	Applications
13:30	Part II
14:19	Data, Functions, Risk
14:42	The Data
17:29	The Function Space
19:34	The Loss Function
21:22	The Risk and the Empirical Risk
24:10	The Risk and the Empirical Risk
26:22	The Training Error
28:03	The Capacity
28:06	The Capacity
32:10	Theoretical Curves
35:26	Theoretical Curves
37:41	Methodology
37:43	Methodology
39:59	Model Selection - Validation
43:05	Model Selection - Cross-validation
46:38	Estimation of the Risk - Validation
47:46	Estimation of the Risk - Cross-validation
47:49	Estimation of the Risk and Model Selection
50:02	Train - Validation - Test
50:05	Cross-validation + Test
50:07	Models
50:08	Examples of Known Models
52:31	Part III
53:01	Preliminaries
53:03	Reminder: Basics on Probabilities
54:28	Gaussian Mixture Models
54:30	What is a Gaussian Mixture Model
56:15	Characteristics of a GMM
57:36	Expectation-Maximization
57:41	Basics of Expectation-Maximization
60:23	EM for GMM (Graphical View, 1)
62:16	EM for GMM (Graphical View, 2)
62:50	EM for GMM (Graphical View, 3)
63:19	EM: More Formally
65:40	EM for GMMs
65:45	EM for GMM: Hidden Variable
67:30	EM for GMM: Auxiliary Function
69:17	EM for GMM: Auxiliary Function
71:01	EM for GMM: Update Rules
72:46	Initialization

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.