Explore the fundamentals of Reinforcement Learning in this comprehensive lecture from the Learning and Games Boot Camp. Delve into key concepts such as credit assignment, dynamic programming, and exploration as presented by Dylan Foster from Microsoft Research. Gain insights into algorithm design, sample complexity, and the UCB algorithm. Examine the connections between reinforcement learning, contextual bandits, and supervised learning. Investigate advanced topics like inverse gap weighting and direct policy search. Enhance your understanding of this crucial field in artificial intelligence and machine learning through this in-depth presentation from the Simons Institute.
Overview
Syllabus
Introduction
Three Problems
Three Steps
Gameplan
Reinforcement Learning
Credit Assignment
Dynamic Programming
Exploration
Algorithm
Generalization
Algorithm Design
Sample Complexity
Roadmap
Ucbvi
Contextual Bandit
Supervised Learning
Inverse Gap Weighting
Direct Policy Search
Taught by
Simons Institute
Related Courses
-
Decision Making and Reinforcement Learning
-
Online Learning and Bandits - Part 2
-
Neural Nets for NLP 2017 - Reinforcement Learning
-
Multi-Agent Reinforcement Learning - Part II
-
Provably Efficient Reinforcement Learning with Linear Function Approximation - Chi Jin
-
Neural Nets for NLP 2019 - Reinforcement Learning
