\( \newcommand{\vect}[1]{\mathbf{#1}} \newcommand{\mat}[1]{\mathbf{#1}} \newcommand{\R}{\mathbb{R}} \newcommand{\E}{\mathbb{E}} \newcommand{\Prob}{\mathbb{P}} \newcommand{\T}{^{\top}} \newcommand{\norm}[1]{\lVert #1 \rVert} \newcommand{\abs}[1]{\lvert #1 \rvert} \newcommand{\ihat}{\mkern1mu\hat{\boldsymbol{\imath}}\mkern2mu} \newcommand{\jhat}{\mkern1mu\hat{\boldsymbol{\jmath}}\mkern2mu} \newcommand{\pd}[2]{\frac{\partial #1}{\partial #2}} \newcommand{\argmax}{\operatorname*{arg\,max}} \newcommand{\argmin}{\operatorname*{arg\,min}} \newcommand{\cdata}[1]{\class{c-data}{#1}} \newcommand{\cmodel}[1]{\class{c-model}{#1}} \newcommand{\ctruth}[1]{\class{c-truth}{#1}} \newcommand{\cgrad}[1]{\class{c-grad}{#1}} \newcommand{\closs}[1]{\class{c-loss}{#1}} \newcommand{\cparam}[1]{\class{c-param}{#1}} \newcommand{\csecond}[1]{\class{c-second}{#1}} \renewcommand{\,}{\mkern3mu} \)

RL02 · Tabular Model-Free RL

A free, illustrated L2 course of 9 modules in the Reinforcement Learning track. Modules include: Monte Carlo prediction and control; Temporal-difference learning; SARSA.

RL02 L2 · Intermediate 9 modules · about 36 lessons Planned · Ring 2

Part of the Reinforcement Learning track. Ships in Ring 2 (Spine) of the roadmap.

Modules

1. Monte Carlo prediction and control
2. Temporal-difference learning
3. SARSA
4. Q-learning
5. Expected SARSA and double learning
6. n-step methods
7. Eligibility traces
8. Planning with Dyna
9. On-policy, off-policy and importance sampling

Signature illustrations

The pictures a learner should still see in their head a year later:

Monte Carlo and temporal-difference backups side by side, cliff-walking paths compared, a Q-table updating, decaying traces.

Status

This course is planned, not written yet. It ships with Ring 2 (Spine).

Every lesson will follow the same twelve-part blueprint with at least three purpose-made figures. Watch the repository to hear when modules ship, or help write and review it.