CS422

Download as PDF

Reasoning, Planning and Learning under Uncertainty

Course (UG/PG)

Undergraduate

Offering Unit/Department

Course Description

A key challenge in many AI systems is being able to handle (reason, plan or learn) problems with uncertainty. For instance, a self-driving car needs to handle imperfect visibility of the world and also uncertainty about the movements of other cars or obstacles. Similarly, in aggregation systems (e.g., Grab, Food Panda), positioning supply (taxis, delivery boys) at the right locations requires handling uncertainty about customer demand. Playing strategic games like Go, Chess etc. requires learning strategy that works against opponents whose strategy is not known in advance. This course will equip students with core concepts and practical experience in doing reasoning, learning and planning in the presence of uncertainty.

Course Learning Outcomes

(1) Identify the probability of a credit card fraud given a certain transaction

(2) Calculate strategy of matching taxis to customers

(3) Developing smart strategies for playing strategic computer or board games

Fundamentals of deep learning and Deep Reinforcement

Discipline-Specific Competencies

Change Management, Computational Modelling, Intelligent Reasoning, Pattern Recognition Systems, Self-Learning Systems

SMU Graduate Learning Outcomes

Disciplinary Knowledge, Critical thinking & problem solving, Innovation and enterprising skills, Collaboration and leadership, Communication, Self-directed learning

Grading Basis

GRD - Graded