Explore key issues in reinforcement learning for non-Markovian environments in this 47-minute ISS Informal Systems Seminar talk by Vivek Borkar from the Department of Electrical Engineering at Indian Institute of Technology Bombay. Delve into parallels with classical stochastic control concepts and examine the objective of approximating conditional laws, drawing inspiration from POMDP control using belief states. Gain insights into tackling challenges posed by non-Markovian settings in reinforcement learning and understand their implications for practical applications.
Overview
Syllabus
Reinforcement Learning for Non-Markovian Environments , Vivek Borkar
Taught by
GERAD Research Center
Related Courses
-
On the Curses of Future and History in Off-policy Evaluation in Non-Markov Environments
-
Partially Observable Reinforcement Learning
-
Reinforcement Learning
-
Reinforcement Learning via Stochastic Control
-
How to Train Your Robot - An Introduction to Reinforcement Learning
-
Recent Progress in Continuous-Time Reinforcement Learning and Regret Analysis
