91社区

91社区.ca / Overview / All Courses
important

Note: 91社区鈥檚 new Course Catalogue will replace the eCalendar. The Course Catalogue is expected to go live the week of April 22nd. When the new site is published, "mcgill.ca/study" will be redirected to the new Course Catalogue website.

Course information on this site is not reflective of offerings for the 2025鈥�2026 academic year. Some irregularities may occur as we move operations to the incoming Course Catalogue.

COMP 579 Reinforcement Learning (4 credits)

Offered by: Computer Science (Faculty of Science)

Overview

Computer Science (Sci) : Bandit algorithms, finite Markov decision processes, dynamic programming, Monte-Carlo Methods, temporal-difference learning, bootstrapping, planning, approximation methods, on versus off policy learning, policy gradient methods temporal abstraction and inverse reinforcement learning.

Terms: Winter 2025

Instructors: Precup, Doina; Pr茅mont-Schwarz, Isabeau (Winter)

  • Prerequisite: A university level course in machine learning such as COMP 451 or COMP 551. Background in calculus, linear algebra, probability at the level of MATH 222, MATH 223, MATH 323, respectively.

Back to top