Most tasks require a person or an automated system to reason--to reach conclusionsbased on available information. The framework of probabilistic graphical models, presented in thisbook, provides a general approach for this task. The approach is mode......more
Most tasks require a person or an automated system to reason--to reach conclusionsbased on available information. The framework of probabilistic graphical models, presented in thisbook, provides a general approach for this task. The approach is model-based, allowing interpretablemodels to be constructed and then manipulated by reasoning algorithms. These models can also belearned automatically from data, allowing the approach to be used in cases where manuallyconstructing a model is difficult or even impossible. Because uncertainty is an inescapable aspectof most real-world applications, the book focuses on probabilistic models, which make theuncertainty explicit and provide models that are more faithful to reality. Probabilistic GraphicalModels discusses a variety of models, spanning Bayesian networks, undirected Markov networks,discrete and continuous models, and extensions to deal with dynamical systems and relational data.For each class of models, the text describes the three fundamental cornerstones: representation,inference, and learning, presenting both basic concepts and advanced techniques. Finally, the bookconsiders the use of the proposed framework for causal reasoning and decision making underuncertainty. The main text in each chapter provides the detailed technical development of the keyideas. Most chapters also include boxes with additional material: skill boxes, which describetechniques; case study boxes, which discuss empirical cases related to the approach described in thetext, including applications in computer vision, robotics, natural language understanding, andcomputational biology; and concept boxes, which present significant concepts drawn from the materialin the chapter. Instructors (and readers) can group chapters in various combinations, from coretopics to more technically advanced material, to suit their particular needs.
Daphne Koller is Professor in the Department of Computer Science at Stanford University. Nir Friedman is Professor in the Department of Computer Science and Engineering at Hebrew University.