PGM

Trying to grasp intuition behind inference on Probabilistic Graphical Model is difficult. This project is an attempt to make the visualization of three different algorithms sum-product, max-product, and max-sum algorithm and also shows an application of Markov Random Field to perform image de-noising, which is an important task in image pre-processing. Example Graph, with their joint distribution:

Here is animation showing intermediate of the following algorithms:

1. Sum-product:

2. Max-product:

3. Max-sum:

Please refer to FinalNB for the complete code.

Here's a link to Presentation created out of this project.

PS: This is a project assignment for course CSE627 at Miami University.