Dr. Rina Dechter

Biography

The Dechter lab conducts research in Automated Reasoning and Artificial Intelligence, specifically focusing on Graphical Models. These graph-based models, such as Bayesian networks, influence diagrams, and Markov decision processes, serve as a central paradigm for knowledge representation and reasoning within Artificial Intelligence and Computer Science. The models are employed to accomplish various tasks in science and engineering, including scheduling, planning, learning, diagnosis, prediction, design, and hardware/software verification in bioinformatics. Many reasoning problems are formally articulated as constraint satisfaction or satisfiability tasks, as well as combinatorial optimization and probabilistic inference issues. Some of these tasks are known to be computationally hard; however, research over the past several decades has yielded a variety of principles and techniques that have significantly advanced the state of the art. The lab's approach focuses on devising methods to understand and exploit tractable reasoning tasks while designing general anytime algorithms that provide solutions with an added provision for improving the quality of solutions over time.

Research Interests