Reinforcement Learning 2:0

Prof. Shalabh Bhatnagar

Department of Computer Science and Automatio, IISc

Syllabus

Introduction to Reinforcement Learning, Multi-armed bandits, Markov decision processes, Dynamic Programming – Value and Policy Iteration Methods, Model-Free Learning Approaches, Monte-Carlo Methods, Temporal Difference Learning, Q-learning, SARSA, Double Q-learning, Value Function Approximation Methods – TD Learning with Linear Function Approximation, Neural Network Architectures, Deep Q-Network Algorithm, Policy Gradient Methods, Actor-Critic Algorithms.

WEEK Topic to be covered

Week 1: Introduction to RL
Week 2: Multi-armed bandits
Week 3: Markov decision processes
Week 4: Markov decision processes and dynamic programming
Week 5: Dynamic programming algorithms
Week 6: Monte-Carlo algorithms
Week 7: Temporal Difference Learning: the full-state case
Week 8: Temporal Difference Learning: full state algorithms
Week 9: RL with linear function aproximation
Week 10: RL with nonlinear function approximation: Deep Q-Networks
Week 11: Policy gradient methods
Week 12: Policy gradient methods

Minimum Qualification

B.Tech in any discipline or B.Sc/BA in Physics/Chemistry/Maths/Statistics/Computer Science/Economics

Pre-requisites

NIL

Who can apply?

Any person from the industry or academic institutions should benefit.