Skip to main content
BitsNotes

Deep Reinforcement Learning

17 lectures

Agents that learn from reward through value methods, policy gradients, and exploration strategies.

01Lecture 01 · Introduction to Reinforcement Learning
02Lecture 02 · Elements of RL and Multi-Armed Bandits
03Lecture 03 · Multi-Armed Bandit: Advanced Topics
04Lecture 04 · Markov Decision Processes
05Lecture 05 · Markov Decision Processes and Dynamic Programming
06Lecture 06 · Dynamic Programming — Value Iteration, Policy Iteration, and Q-Learning
07Lecture 07 · Monte Carlo Methods
08Lecture 08 · Monte Carlo Methods: On-Policy Prediction and Control
09Lecture 09 · Off-Policy Monte Carlo Methods and Introduction to Temporal Difference Learning
10Lecture 10 · Temporal Difference Learning: TD(0), SARSA, and Q-Learning
11Lecture 11 · Temporal Difference Learning, Maximization Bias, and Function Approximation
12Lecture 12 · Feature Construction and Deep Q-Networks (DQN)
13Lecture 13 · Deep Q-Networks and Policy Gradient Methods
14Lecture 14 · Policy Gradient Methods
15Lecture 15 · Model-Based Learning and Monte Carlo Tree Search
16Lecture 16 · Comprehensive Examination Revision and Advanced Policy Methods
17Lecture 17 · Imitation Learning
Loading comments…