Tabular RL + Classic Control
The cleanest path for reviewing how RL pieces fit together before deep networks dominate the picture.
How It Works
This branch is notebook-first. It walks from policy/value iteration into Monte Carlo and temporal-difference learning, then moves into discretization, tile coding, hill climbing, cross-entropy, and Taxi. The material is structured as short exercises plus matching solution notebooks.
Why Reuse
These notebooks are the best first candidates for live DRL web lessons because they explain the math cleanly and usually avoid Unity-specific baggage.
Key Assets
Policy evaluation, improvement, iteration, and value iteration.
source-material/dynamic-programming
Prediction and control notebook pair with plotting helpers.
source-material/monte-carlo
Sarsa, Q-learning, and expected Sarsa progression.
source-material/temporal-difference
Continuous-state discretization bridge into function approximation.
source-material/tile-coding
Self-contained small environment with `agent.py`, `monitor.py`, and `main.py`.
source-material/taxi