Planning, Search, and Reasoning Under Uncertainty

Catalog description

This course delves into the exploration of defining planning domains, which includes creating representations for world states and actions, encompassing both symbolic and path planning. Students will examine algorithms aimed at efficiently generating valid plans, whether optimized or not, and providing either partially ordered or fully specified solutions. Furthermore, delve into decision-making processes and Number Name Credits their practical applications in addressing real-world challenges involving complex autonomous systems. The exploration will also focus on efficient search methods for finding solutions in planning domains with finite state lengths. Moreover, to enhance eff ectiveness in real -world planning and action, study methods for reasoning about sensing, actuation, and handling model uncertainty. Throughout the course, draw connections between classical approaches that offered initial solutions to these challenges and modern machine learning techniques, showcasing how they build upon and complement traditional methodologies.