feat: execute v5 Phase-I roadmap (V1~V5)#65
Merged
Conversation
V1 — R15 완성: utils/hp_search.py CVaR-aware 격자 탐색
- CVaRAwareHPSearch: MonteCarloEvaluator 기반 HP 탐색
- Pareto 프런트(mean_win_rate vs cvar_10) 계산 로직
- CVaR 배포 게이트(min_win_rate + max_cvar10_casualties) 필수 조건화
- ExperimentRegistry 자동 등록 + JSON 저장
- tests/test_hp_search.py: 7개 테스트 (config, 인스턴스화, 조합수, Pareto, 폴백, e2e, summary)
V2 — AdaptiveMCDropout GNN 통합
- BayesianHGT.__init__에 adaptive_mc=True 옵션 추가
- predict_with_uncertainty()에 AdaptiveMCDropout 경로 추가
- 조기 수렴 시 _last_mc_n_runs 기록 (로깅/디버깅용)
- 고정 mc_samples 경로 유지 (adaptive_mc=False 폴백)
V3 — sys.path.insert 패키지 내부 제거
- evaluation/monte_carlo.py: _mc_run_worker()의 sys.path.insert 제거
(ProcessPoolExecutor가 부모 sys.path 상속, pyproject.toml pythonpath로 충분)
- evaluation/adversarial_benchmark.py: __main__ 블록에 중복 방지 가드 추가
V4 — CF 피드백 루프 Phase 3·4 확장
- train.py: _setup_cf_loop() 공통 헬퍼 추가 (Phase 1·3·4 재사용)
- Phase 1: 기존 CF 생성 코드를 _setup_cf_loop() 호출로 교체 (동작 동일)
- Phase 3: cf_loop_p3 활성화 — HITL 선택 이벤트를 CF 에피소드로 기록
- Phase 4: cf_loop_p4 활성화 — PPO 버퍼 보상 보정 전 Phase 1과 동일 흐름
V5 — TierCTrainer 평가 경로 통합
- TierCTrainer.select_action() 공개 메서드 추가 (BlueAgent 호환 인터페이스)
- evaluate_vs_baseline(): 내부 경량 루프 → MonteCarloEvaluator 직접 사용
- 비교 지표에 cvar_10, delta_cvar_10, strategy_robustness, win_rate_ci 추가
- Tier A(PPO)와 완전히 동일한 평가 경로 보장
https://claude.ai/code/session_013zzx6Mwbd5LJacDUsNbitA
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.