Introduction to Probabilistic Graphical Models

Speaker:  Indrajit Bhattacharya – Kolkata, India
Topic(s):  Artificial Intelligence, Machine Learning, Computer Vision, Natural language processing

Abstract

Probabilistic Graphical Models (PGM) provide a unifying framework for representing and reasoning with complex domains with many uncertain and interdependent variables and form one of the cornerstones of artificial intelligence. A wide range of models and algorithms such as Naive Bayes, Mixture Models, PCA, ICA, Bayesian and Markov Networks, Hidden Markov Models, Conditional Random Fields, Linear Dynamical Systems (including Kalman Filters), Boltzmann Machines, Hierarchical Bayesian Models can be analyzed within this framework. In this lecture, I will talk about the theoretical foundations of PGMs including directed and undirected graphical models considering Hidden Markov Models and Conditional Random Fields as examples for the task of text analysis. I will describe the hardness of inference using PGMs and outline variable elimination and belief propagation as algorithms for inference. I will also touch upon the problem of parameter estimation and structure learning in PGMs. 

About this Lecture

Number of Slides:  120
Duration:  120 - 180 minutes
Languages Available:  English
Last Updated: 

Request this Lecture

To request this particular lecture, please complete this online form.

Request a Tour

To request a tour with this speaker, please complete this online form.

All requests will be sent to ACM headquarters for review.