Protocol
Reinforcement Learning from Bellman Equations
Build an RL study with a known tabular oracle before a neural agent. Analyze one reward or exploration failure and propose a controlled correction.
60–100 active hours60–90 min sessionsAdvanced
Mastery contract
Specialization · Standard 1.0 · Build an RL study with a known tabular oracle before a neural agent. Analyze one reward or exploration failure and propose a controlled correction.
Loading your learning records…
Reasoning — not yet passed
Derive both Bellman equations and the score-function gradient without mixing state value and action value.
Reliable implementation — not yet passed
Tabular values agree with exact dynamic programming within tolerance; terminal transitions are tested.
Reproducible experiment — not yet passed
Compare the learned policy with a declared random or tabular baseline under matched environment/sample budgets; retain return distributions, all seeds and failure rates.
Defense and handoff — not yet passed
Defend the evaluation and reward design, identifying a behavior that exploits the reward without solving the intended task.
All four criteria must pass. Activity completion does not satisfy them.
An independent reviewer reruns your work and varies something you did not rehearse: a fresh input, a different working directory, or a declared edge case.
Pilot decisions are provisional, not external credentials. For an appeal or sensitive artifact, contact your designated pilot operator with the submission ID. Export your assessment history.
What you will do
6 total
Derive Bellman expectation/optimality equations and the REINFORCE gradient for a finite-horizon episodic setting; state the assumptions.
Implement value iteration and tabular Q-learning on a tiny MDP with a known optimum, then a small policy-gradient agent.
Compare exploration settings over at least 10 seeds with fixed episode budgets; evaluate greedy policies separately from training returns.
Debug terminal-state bootstrapping and a reward that encourages an unintended behavior; distinguish truncation from actual termination.
Write a 500–900 word technical report linking the reasoning, code, measurements, and limitations; include the project decision below.
Assess all 4 mastery criteria against saved artifacts, request an independent review, and repeat each failed criterion on a fresh example.
Essentials
A reading session alone never satisfies a mastery criterion.
Store source references, assumptions, and artifact paths beside every result.
Keep an untouched check case that differs from the worked example.
Record all attempts, including failures and results that contradict your prediction.
Use the stated comparison conditions; document every deviation before drawing a conclusion.
Protocol guardrails
Do
- +State the expected result before running the comparison.
- +Keep one minimal reproducible failing case when debugging.
- +Record environment versions and the exact command used.
- +Compare explanations with saved intermediate values.
- +Ask a reviewer to challenge the weakest assumption.
Don't
- ×Do not copy a worked solution and present it as an independent implementation.
- ×Do not tune against held-out evaluation outcomes.
- ×Do not report only the best seed or discard inconvenient runs.
- ×Do not equate elapsed hours or a completed run with a passed assessment.
- ×Do not conceal reduced-scale experiments behind claims about the original full-scale result.
Protocol authorship
Written by NuthinButta as instructional design. The teaching sources are the official references linked in each milestone; the exercises, workload and pass thresholds are ours, not their authors'.