Large Scale Graphical Models

Large Scale Graphical Models

8 Lectures · Dec 12, 2009

About

Approximate Learning of Large Scale Graphical Models: Theory and Applications

Undirected graphical models provide a powerful framework for representing dependency structure between random variables. Learning the parameters of undirected models plays a crucial role in solving key problems in many machine learning applications, including natural language processing, visual object recognition, speech perception, information retrieval, computational biology, and many others. Learning in undirected graphical models of large treewidth is difficult because of the hard inference problem induced by the partition function for maximum likelihood learning, or by finding the MAP assignment for margin-based loss functions. Over the last decade, there has been considerable progress in developing algorithms for approximating the partition function and MAP assignment, both via variational approaches (e.g., belief propagation) and sampling algorithms (e.g., MCMC). More recently, researchers have begun to apply these methods to learning large, densely-connected undirected graphical models that may contain millions of parameters. A notable example is the learning of Deep Belief Networks and Deep Boltzmann Machines, that employ MCMC strategy to greedily learn deep hierarchical models.

The Workshop homepage can be found at http://www.cs.toronto.edu/~rsalakhu/workshop_nips2009/index.html.

Related categories

Uploaded videos:

video-img
33:05

Approximate Inference in Natural Language Processing

Noah Smith

Jan 19, 2010

 · 

5098 Views

Lecture
video-img
34:50

Learning Deep Boltzmann Machines

Ruslan Salakhutdinov

Jan 19, 2010

 · 

17318 Views

Lecture
video-img
41:56

Joint Max Margin and Max Entropy Learning of Graphical Models

Eric P. Xing

Jan 19, 2010

 · 

5303 Views

Lecture
video-img
43:32

Large-Scale Learning and Inference: What We Have Learned with Markov Logic Netwo...

Pedro Domingos

Jan 19, 2010

 · 

7030 Views

Lecture
video-img
17:04

Parameter Learning Using Approximate MAP Inference

Pawan Kumar Mudigonda

Jan 19, 2010

 · 

3986 Views

Lecture
video-img
11:03

Learning a Region-based Scene Segmentation Model

Pawan Kumar Mudigonda

Jan 19, 2010

 · 

3272 Views

Lecture
video-img
28:56

Training Structured Predictors for Novel Loss Functions

David McAllester

Jan 19, 2010

 · 

3462 Views

Lecture
video-img
30:16

Some Machine Learning Problems that We in the Computer Vision Community Would Li...

William T. Freeman

Jan 14, 2010

 · 

6538 Views

Lecture