Explore

Probabilistic Graphical Models 2: Inference

Approx. 38 hours to complete

Save Course

Go to Course

Course Summary

Learn how to build probabilistic graphical models to make predictions and decisions in complex systems. This course covers inference in graphical models, exact and approximate algorithms for inference, and methods for learning graphical models from data.

Key Learning Points

Understand the basics of probabilistic graphical models
Learn about inference algorithms and exact inference methods
Explore approximate inference methods for large and complex models
Discover techniques for learning graphical models from data

Learning Outcomes

Develop expertise in building probabilistic graphical models
Gain knowledge of inference algorithms and exact inference methods for graphical models
Learn the techniques for learning graphical models from data

Prerequisites or good to have knowledge before taking this course

Basic knowledge of probability theory and linear algebra
Familiarity with Python programming

Course Difficulty Level

Intermediate

Course Format

Self-paced
Online

Similar Courses

Probabilistic Graphical Models
Bayesian Statistics: From Concept to Data Analysis
Applied Data Science with Python

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 second in a sequence of three. Following the first course, which focused on representation, this course addresses the question of probabilistic inference: how a PGM can be used to answer questions. Even though a PGM generally describes a very high dimensional distribution, its structure is designed so as to allow questions to be answered efficiently. The course presents both exact and approximate algorithms for different types of inference tasks, and discusses where each could best be applied. The (highly recommended) honors track contains two hands-on programming assignments, in which key routines of the most commonly used exact and approximate algorithms are implemented and applied to a real-world problem.

Outline

Inference Overview
Overview: Conditional Probability Queries
Overview: MAP Inference

Variable Elimination
Variable Elimination Algorithm
Complexity of Variable Elimination
Graph-Based Perspective on Variable Elimination
Finding Elimination Orderings
Variable Elimination

Belief Propagation Algorithms
Belief Propagation Algorithm
Properties of Cluster Graphs
Properties of Belief Propagation
Clique Tree Algorithm - Correctness
Clique Tree Algorithm - Computation
Clique Trees and Independence
Clique Trees and VE
BP In Practice
Loopy BP and Message Decoding
Message Passing in Cluster Graphs
Clique Tree Algorithm

MAP Algorithms
Max Sum Message Passing
Finding a MAP Assignment
Tractable MAP Problems
Dual Decomposition - Intuition
Dual Decomposition - Algorithm
MAP Message Passing

Sampling Methods
Simple Sampling
Markov Chain Monte Carlo
Using a Markov Chain
Gibbs Sampling
Metropolis Hastings Algorithm
Sampling Methods
Sampling Methods PA Quiz

Inference in Temporal Models
Inference in Temporal Models
Inference in Temporal Models

Inference Summary
Inference: Summary
Inference Final Exam

Summary of User Reviews

Key Aspect Users Liked About This Course

Many users found the course to be well-structured and informative, providing a deep understanding of probabilistic graphical models and their applications.

Pros from User Reviews

Excellent course material with clear explanations and practical examples.
The instructor is knowledgeable and engaging, making the course enjoyable and easy to follow.
Great for those who want to enhance their understanding of probabilistic graphical models and their applications.
The assignments and quizzes are challenging but rewarding, allowing learners to apply their knowledge in practice.

Cons from User Reviews

Some users found the course to be too advanced and challenging for beginners.
The course requires a solid understanding of probability theory and linear algebra.
Some users felt that the course could have included more practical applications and real-world examples.
The course can be time-consuming and demanding, requiring a significant commitment from learners.
Some users experienced technical difficulties with the Coursera platform or the course materials.

Recommended for you

Probability - The Science of Uncertainty and Data

The world is full of uncertainty: accidents, storms, unruly financial markets, noisy communications. The world is also full of data. Probabilistic models use the language of mathematics....

Save Course

Introduction to Probability: Part II – Inference & Processes

Learn how to use probability theory to develop the basic elements of statistical inference and important random process models The world is also full of data....

Save Course