Agents, RL & Decision Science

Step 7 Non Linear Models Examples 7.6 Agent Based Modeling

A generic course about Agent Based Modeling. Content coming soon.

Decision Theory & Robust Preferences

Foundations of rational decision-making: utility theory, risk measures, and robust preference models.

Online Learning & Adversarial Bandits

Regret minimization in online learning: experts, adversarial bandits, and multiplicative weights.

Contextual Bandits & Off-Policy Evaluation

Contextual bandits for personalization with off-policy evaluation methods for safe deployment.

Policy Learning & Counterfactual Risk Minimization

Learn optimal policies from logged bandit data using counterfactual risk minimization.

MDPs & Dynamic Programming

Markov decision processes: Bellman equations, value iteration, and policy iteration foundations.

RL with Function Approximation

Reinforcement learning with linear and neural function approximation: DQN, policy gradient, and convergence analysis.

Safe, Robust & Risk-Sensitive RL

RL under safety constraints: constrained MDPs, robust MDPs, and risk-sensitive objectives.

Inverse RL & Imitation Learning

Learn reward functions from demonstrations: IRL, behavioral cloning, and DAgger.

Off-Policy Evaluation: IS, DR, FQE Guarantees

Theory of off-policy evaluation in RL: importance sampling, doubly robust methods, and fitted Q evaluation.

Sample Complexity & PAC-style Guarantees in RL

PAC-MDP framework, sample complexity bounds, and minimax rates for reinforcement learning.

POMDPs & Information-State Control

Partially observable MDPs: belief states, information states, and planning under partial observability.