Explore

Probabilistic Graphical Models 3: Learning

Approx. 66 hours to complete

Save Course

Go to Course

Course Summary

Learn about probabilistic graphical models and their applications in machine learning, computer vision, and natural language processing. This course covers the basics of probabilistic graphical models, including Bayesian and Markov networks, and how to use them for prediction and inference.

Key Learning Points

Understand the basics of probabilistic graphical models and their applications
Learn how to use Bayesian and Markov networks for prediction and inference
Explore real-world applications of probabilistic graphical models in machine learning

Learning Outcomes

Develop a strong understanding of probabilistic graphical models
Be able to apply Bayesian and Markov networks for prediction and inference
Learn how to use probabilistic graphical models in real-world applications

Prerequisites or good to have knowledge before taking this course

Basic knowledge of probability theory and linear algebra
Experience with programming in Python

Course Difficulty Level

Intermediate

Course Format

Online
Self-paced

Similar Courses

Probabilistic Graphical Models
Probabilistic Graphical Models

Related Education Paths

Related Books

Description

Probabilistic graphical models (PGMs) are a rich framework for encoding probability distributions over complex domains: joint (multivariate) distributions over large numbers of random variables that interact with each other. These representations sit at the intersection of statistics and computer science, relying on concepts from probability theory, graph algorithms, machine learning, and more. They are the basis for the state-of-the-art methods in a wide variety of applications, such as medical diagnosis, image understanding, speech recognition, natural language processing, and many, many more. They are also a foundational tool in formulating many machine learning problems.

This course is the third in a sequence of three. Following the first course, which focused on representation, and the second, which focused on inference, this course addresses the question of learning: how a PGM can be learned from a data set of examples. The course discusses the key problems of parameter estimation in both directed and undirected models, as well as the structure learning task for directed models. The (highly recommended) honors track contains two hands-on programming assignments, in which key routines of two commonly used learning algorithms are implemented and applied to a real-world problem.

Outline

Learning: Overview
Learning: Overview

Review of Machine Learning Concepts from Prof. Andrew Ng's Machine Learning Class (Optional)
Regularization: The Problem of Overfitting
Regularization: Cost Function
Evaluating a Hypothesis
Model Selection and Train Validation Test Sets
Diagnosing Bias vs Variance
Regularization and Bias Variance

Parameter Estimation in Bayesian Networks
Maximum Likelihood Estimation
Maximum Likelihood Estimation for Bayesian Networks
Bayesian Estimation
Bayesian Prediction
Bayesian Estimation for Bayesian Networks
Learning in Parametric Models
Bayesian Priors for BNs

Learning Undirected Models
Maximum Likelihood for Log-Linear Models
Maximum Likelihood for Conditional Random Fields
MAP Estimation for MRFs and CRFs
Parameter Estimation in MNs

Learning BN Structure
Structure Learning Overview
Likelihood Scores
BIC and Asymptotic Consistency
Bayesian Scores
Learning Tree Structured Networks
Learning General Graphs: Heuristic Search
Learning General Graphs: Search and Decomposability
Structure Scores
Tree Learning and Hill Climbing

Learning BNs with Incomplete Data
Learning With Incomplete Data - Overview
Expectation Maximization - Intro
Analysis of EM Algorithm
EM in Practice
Latent Variables
Learning with Incomplete Data
Expectation Maximization

Learning Summary and Final
Summary: Learning
Learning: Final Exam

PGM Wrapup
PGM Course Summary

Summary of User Reviews

Learn about probabilistic graphical models in this course. Users have generally positive reviews about this course with many praising the depth of the material covered.

Key Aspect Users Liked About This Course

Many users found the course content to be very comprehensive and in-depth.

Pros from User Reviews

The course covers a wide range of topics in probabilistic graphical models.
The instructors are knowledgeable and provide clear explanations.
The assignments and quizzes are challenging but rewarding.
The course provides ample opportunities for hands-on learning.
The video lectures are well-produced and engaging.

Cons from User Reviews

The course may be too advanced for beginners in the subject.
Some users found the pace of the course to be too fast.
The programming assignments require a significant amount of time and effort.
The course may not be suitable for those looking for a more theoretical approach.
The course may require additional resources for a deeper understanding of the material.

Recommended for you

Complete Linear Regression Analysis in Python

Linear Regression in Python| Simple Regression, Multiple Regression, Ridge Regression, Lasso and subset selection also You've found the right Linear Regression course!...

Save Course

Logistic Regression in Python

Logistic regression in Python tutorial for beginners. You can do Predictive modeling using Python after this course. You've found the right Classification modeling course!...

Save Course

Probabilistic Graphical Models 1: Representation

Save Course

Text Retrieval and Search Engines

This course will cover search engine technologies, which play an important role in any data mining applications involving text data for two reasons....

Save Course