May 6, 2014
Professor Rina Dechter recently published a new book based upon her extensive work in artificial intelligence with application to machine learning, titled Reasoning with Probabilistic and Deterministic Graphical Models: Exact Algorithms. Published by Morgan Claypool in December, the work is part of the firm’s series, Synthesis Lectures on Artificial Intelligence and Machine Learning.
In the book, Dechter, a computer science professor and vice chair of her department’s computing division, describes how computation can be performed with graphical models (e.g., Bayesian and constraint networks, influence diagrams, and Markov decision processes), which have become a central paradigm for knowledge representation and reasoning in both artificial intelligence in particular and computer science in general. Such tasks are computationally difficult, but Dechter states that research during the past three decades has yielded a variety of principles and techniques that have significantly advanced the state of the art.
Dechter goes on to provide comprehensive coverage of the primary exact algorithms for reasoning with such models. The main feature exploited by the algorithms is the model's graph. She presents inference-based message-passing schemes and search-based, conditioning schemes. Dechter writes that the principles outlined in the book would serve well in moving forward to approximation and anytime-based schemes. The target audience of the book is researchers and students in the artificial intelligence and machine learning area — and beyond.
Dechter’s research into the algorithmic foundations of automated reasoning with constraint-based and probabilistic information helped earn her a prestigious Association of Computing Machinery (ACM) fellowship late last year. She also is the author of the book Constraint Processing (2003) and has authored more than 150 papers.
UCI ICS Feature: Book it