Probabilistic Graphical Models 2 Inference at Stanford

Probabilistic Graphical Models (PGMs) is considered to be a strong foundation for encrypting probability arrangement over composite domains. Hence, Probabilistic Graphical Models 2: Inference is drafted to address the questions related to probabilistic inference. This course depends upon the concepts from graph algorithms, probability theory, and machine learning. These concepts are the base for the advanced methods used in various applications like speech recognition, medical diagnosis, natural language processing, etc. These advanced methods are also essential tools in composing various machine learning problems.

In an order of three, Probabilistic Graphical Models 2: Inference is the second one. While the first course, focused on description, this course would address the question occurs in probabilistic inference or in other words use of PGM in answering the questions. PGM is described as a very high dimensional diffusion but its framework is drafted in such a way that it can answer the questions efficiently. The course introduces both approximate and accurate algorithms for the variety of inference tasks, and also talks over the application of each algorithm.