Probabilistic Graphical Models (PGMs) are a framework for representing complex distributions over random variables using graphs. In these models, nodes represent random variables, while edges indicate dependencies between them. This structure allows for efficient computation and inference, making it easier to understand relationships in data. PGMs can be classified into two main types: Bayesian Networks and Markov Random Fields. Bayesian Networks use directed edges to represent causal relationships, while Markov Random Fields use undirected edges to capture symmetric relationships. Both types are widely used in fields like machine learning, computer vision, and bioinformatics for tasks such as classification and prediction.