Protocol
Adversarial Generators and Failure Analysis
Use an exactly known multimodal target to investigate collapse. The deliverable is a failure analysis, not a gallery of cherry-picked samples.
35–65 active hours60–90 min sessionsAdvanced
Mastery contract
Specialization · Standard 1.0 · Use an exactly known multimodal target to investigate collapse. The deliverable is a failure analysis, not a gallery of cherry-picked samples.
Loading your learning records…
Reasoning — not yet passed
Derive the optimal discriminator on a tractable distribution, check its value numerically, and identify an optimization assumption broken by the implemented game.
Reliable implementation — not yet passed
Parameter-isolation tests and gradient checks pass on the toy generator/discriminator.
Reproducible experiment — not yet passed
Report mode coverage and failures for all 10 runs with matched baseline settings, seeds, raw counts and compute budgets; selected sample images cannot establish coverage.
Defense and handoff — not yet passed
Use saved training trajectories and a controlled intervention to explain a failure, separating equilibrium theory from observed empirical stability.
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 the optimal discriminator for fixed generator distributions and explain why that result does not guarantee finite-network training convergence.
Implement a GAN on a synthetic mixture with known modes; test that discriminator updates do not update generator parameters and vice versa.
Compare 2 generator objectives over 5 seeds with identical budgets; report mode coverage, sample locations and both losses.
Debug detached generator gradients and mode collapse; introduce a control to distinguish plotting artifacts from genuine missing modes.
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'.