Reinforcement Learning

MDPs, policy vs value methods, exploration strategies, and RL algorithms

Foundations

Reinforcement Learning
Markov Decision Processes
An MDP formalizes sequential decision-making: an agent in a state takes an action, receives a reward, and transitions to a new state, with the goal of maximizing cumulative discounted reward over time.

Value Based Methods

Reinforcement Learning
Bellman Equations
This concept covers the fundamentals of bellman equations within the broader context of Reinforcement Learning.
Reinforcement Learning
Value Functions
This concept covers the fundamentals of value functions within the broader context of Reinforcement Learning.
Reinforcement Learning
Q-Learning
Q-learning estimates the value of taking each action in each state without needing a model of the environment, converging to the optimal action-value function by bootstrapping off its own past estimates.
Reinforcement Learning
Deep Q Networks
This concept covers the fundamentals of deep q networks within the broader context of Reinforcement Learning.
Reinforcement Learning
Epsilon Greedy
This concept covers the fundamentals of epsilon greedy within the broader context of Reinforcement Learning.
Reinforcement Learning
Reward Shaping
This concept covers the fundamentals of reward shaping within the broader context of Reinforcement Learning.
Reinforcement Learning
Temporal Difference Learning
This concept covers the fundamentals of temporal difference learning within the broader context of Reinforcement Learning.
Reinforcement Learning
Monte Carlo Methods
This concept covers the fundamentals of monte carlo methods within the broader context of Reinforcement Learning.
Reinforcement Learning
Model Based
This concept covers the fundamentals of model based within the broader context of Reinforcement Learning.
Reinforcement Learning
Multi Armed Bandits
This concept covers the fundamentals of multi armed bandits within the broader context of Reinforcement Learning.
Reinforcement Learning
Hierarchical
This concept covers the fundamentals of hierarchical within the broader context of Reinforcement Learning.
Reinforcement Learning
Inverse
This concept covers the fundamentals of inverse within the broader context of Reinforcement Learning.
Reinforcement Learning
Imitation Learning
This concept covers the fundamentals of imitation learning within the broader context of Reinforcement Learning.
Reinforcement Learning
Multi Agent
This concept covers the fundamentals of multi agent within the broader context of Reinforcement Learning.
Reinforcement Learning
Safe
This concept covers the fundamentals of safe within the broader context of Reinforcement Learning.
Reinforcement Learning
Rl From Human Feedback
This concept covers the fundamentals of rl from human feedback within the broader context of Reinforcement Learning.

Policy Based Methods

Core Concepts

Reinforcement Learning
Exploration vs. Exploitation
An RL agent must balance trying known good actions (exploitation) against trying new actions that might be better (exploration) — too little exploration gets stuck in local optima, too much wastes time on bad actions.