coolguyandClaude Opus 4.8 774faf2b51 Add the altair-vs-borealis measurement driver ...

The script behind runs/jax-ppo-match/altair_vs_borealis.json: duplicate,
seat-swapped play between the two models across single-deal and three-round
conditions, reporting win rate with a Wilson interval and mean margin. Kept so the
2x2 result is reproducible.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XBQKgvBbxbheiTF1AVy1Sh

2026-07-15 18:11:27 +09:00

..

ablate_match.py

Ablate the match stack: both-seat training carries it, the privileged critic hurts

2026-07-15 03:49:24 +09:00

altair_vs_borealis.py

Add the altair-vs-borealis measurement driver

2026-07-15 18:11:27 +09:00

analyze_first_open_advantage.py

Rename bot family, curate analyze plots, tier evaluation cadence

2026-05-10 15:32:55 +09:00

analyze_first_open_counterfactual.py

Rename bot family, curate analyze plots, tier evaluation cadence

2026-05-10 15:32:55 +09:00

analyze_first_open_followup.py

Rename bot family, curate analyze plots, tier evaluation cadence

2026-05-10 15:32:55 +09:00

analyze_first_open_targets.py

Honour all_negative_fallback in interleaved scheduler; sync default.yaml

2026-05-10 15:01:46 +09:00

bench_amp_trainer.py

Wire AMP into Deep CFR trainer behind run.use_amp flag (default off)

2026-05-07 20:21:17 +09:00

bench_inference_backend.py

Focus project on JAX PPO

2026-07-14 20:09:03 +09:00

check-python.sh

Python 포맷팅과 pre-commit 설정

2026-05-06 19:59:07 +09:00

diminishing_returns.py

Add diminishing returns diagnostic report

2026-07-05 18:32:47 +09:00

eval_current_vs_average.py

Rename bot family, curate analyze plots, tier evaluation cadence

2026-05-10 15:32:55 +09:00

export_jax_ppo_onnx.py

Focus project on JAX PPO

2026-07-14 20:09:03 +09:00

export_match_onnx.py

Ship borealis to the browser, and add the classic three-round mode

2026-07-15 07:06:29 +09:00

export_safe_heuristic_snapshots.py

Rename bot family, curate analyze plots, tier evaluation cadence

2026-05-10 15:32:55 +09:00

final_cycles.py

Add final cycle report and human play CLI

2026-07-06 00:14:24 +09:00

format-python.sh

Python 포맷팅과 pre-commit 설정

2026-05-06 19:59:07 +09:00

generate_match_parity_fixture.py

Ship borealis to the browser, and add the classic three-round mode

2026-07-15 07:06:29 +09:00

generate_web_parity_fixture.py

Add on-device web client

2026-07-14 19:49:03 +09:00

librarian_check_citations.py

Add librarian Stage 1 citation checker with ignore list

2026-05-07 23:22:34 +09:00

librarian_check_links.py

Bootstrap librarian Stage 1 with lychee link checker

2026-05-07 23:15:06 +09:00

librarian_check_memory_drift.py

Complete librarian Stage 1: stale plans + MEMORY drift

2026-05-07 23:44:25 +09:00

librarian_check_oversize.py

Add librarian Stage 1 oversize check

2026-05-07 23:31:37 +09:00

librarian_check_stale_plans.py

Complete librarian Stage 1: stale plans + MEMORY drift

2026-05-07 23:44:25 +09:00

librarian_promote.py

Accept first survey batch + add commit-hash post-processing

2026-05-08 01:52:08 +09:00

librarian_survey.py

Close librarian: full archive promote-survey + parallel dispatch

2026-05-08 02:33:51 +09:00

librarian-ignore.txt

Focus project on JAX PPO

2026-07-14 20:09:03 +09:00

librarian-prompt.md

Bootstrap librarian Stage 1 with lychee link checker

2026-05-07 23:15:06 +09:00

librarian.sh

Complete librarian Stage 1: stale plans + MEMORY drift

2026-05-07 23:44:25 +09:00

measure_exploitability.py

Correct the exploiter budget: it was a quarter of the target, not half

2026-07-15 05:50:58 +09:00

profile_gpu_forward.py

Focus project on JAX PPO

2026-07-14 20:09:03 +09:00

render_jax_ppo_transcripts.py

Add JAX PPO ladder verification pass

2026-07-05 00:17:49 +09:00

serve_web_with_logs.py

Ship borealis to the browser, and add the classic three-round mode

2026-07-15 07:06:29 +09:00

train_match.py

Run the classic self-play trainer and record what the carry probe says

2026-07-15 02:44:04 +09:00