| Fill-conditioned tables warm-started from the frozen n-tuple leaf, beside an unconditioned continuation and two no-training edits of the frozen tables, screened once on a fresh 512-game block at depth 3 and depth 4 RS-20260906T234914Z-a3fae1a9 | valid runoutcome: fail | tier: public-development | | Row and column lookup tables, learned from a billion moves |
| The frozen n-tuple tables as the leaf of the depth-4 fair search: a two-by-two leaf-by-depth screen on a fresh block RS-20260906T171746Z-1623f833 | valid runoutcome: pass | tier: public-development | | Row and column lookup tables, learned from a billion moves |
| Fresh-block replication of the row-and-column n-tuple leaf at larger scale: wider tables trained until the validation margins plateau, screened beside the first candidate RS-20260906T040113Z-6ba93171 | valid runoutcome: pass | tier: public-development | | Row and column lookup tables, learned from a billion moves |
| Row-and-column n-tuple network trained by temporal-coherence TD on the Rust engine, screened as the leaf of the depth-3 seven-stratum fair search RS-20260905T215332Z-95d18a5a | valid runoutcome: pass | tier: public-development | | Row and column lookup tables, learned from a billion moves |
| Rust exact leaf and cache efficiency checks RS-20260905T193830Z-9733627a | valid runoutcome: pass | tier: mechanics-only | | Rust bitboard engine |
| Second continuation of the depth-5-distilled NNUE leaf evolution: resume from the plateau population with a slower mutation-size decay (up to 1,000 further generations) RS-20260904T090620Z-e5731bf0 | valid runoutcome: fail | tier: public-development | | An evolved NNUE leaf, distilled from a depth-5 teacher |
| Continued whole-game evolution of the depth-5-distilled NNUE leaf from the first run's generation-60 population, with annealed mutation and a preregistered plateau stop (up to 1,000 generations) RS-20260903T163321Z-733076b5 | valid runoutcome: fail | tier: public-development | | An evolved NNUE leaf, distilled from a depth-5 teacher |
| Depth-5-distilled NNUE leaf refined by paired-fitness whole-game evolution inside the depth-3 fair search (v2: operational parameters fixed in the record body) RS-20260903T025751Z-6577b33e | valid runoutcome: fail | tier: public-development | | An evolved NNUE leaf, distilled from a depth-5 teacher |
| Linear leaf refit to the depth-4 root value: held-out fit, depth-5 agreement proxy, and gameplay pilot RS-20260903T013022Z-b0937bf9 | valid runoutcome: fail | tier: pilot | | One-ply Q as a pruning prior and a refit leaf |
| Pruned fair search against full-width fair d4s7 on the previously evaluated 256-game pilot cohort RS-20260903T013022Z-d32ee053 | valid runoutcome: inconclusive | tier: pilot | | One-ply Q as a pruning prior and a refit leaf |
| Depth-4 sibling-value panel, linear Q fit, and pruned-search regret on held-out roots RS-20260902T202705Z-75d87947 | valid runoutcome: fail | tier: pilot | | One-ply Q as a pruning prior and a refit leaf |
| Port the six-feature linear Q-learner onto the Rust bitboard engine: parity gates and a diagnostic pilot on previously evaluated development seeds RS-20260902T084356Z-2488ecc7 | valid runoutcome: pass | tier: pilot | | Klein-Friedmann linear Q-learning |
| Cross-entropy search over the six feature weights against whole-game lifetime, frozen and read once on the pilot cohort RS-20260902T084356Z-784ebf14 | valid runoutcome: pass | tier: pilot | | Klein-Friedmann linear Q-learning |
| Reproduce the Klein-Friedmann Q-learning report in its own simulator at pinned commit 8cc8a0e RS-20260902T082726Z-75606ce7 | valid runoutcome: pass | tier: mechanics-only | | Klein-Friedmann linear Q-learning |
| Central frontier scheduler parity and within-decision scaling check RS-20260825T052959Z-1b3ed9a5 | valid runoutcome: pass | tier: mechanics-only | | Rust bitboard engine |
| Rust search-matrix analytics and EC2 launch reproducibility check RS-20260825T052959Z-57698687 | valid runoutcome: pass | tier: mechanics-only | | Rust bitboard engine |
| Rust bitboard engine: parity gates and throughput benchmark vs TypeScript and C++ engines RS-20260824T075451Z-e89ea128 | valid runoutcome: pass | tier: mechanics-only | | Rust bitboard engine |
| E-FAST-M6: port M=6 reveal sampling into the fast memo engine, trace-equivalent to the native factored search RS-20260824T010000Z-8f3e9b4f | valid runoutcome: pass | tier: mechanics-only | | Separate reveal and next-disc sampling |
| P-SOL-2: corrected two-rung G0 fidelity ladder (fast-d3s7 proxy + native D3 N7M6 guardrail), fast-engine corpus continuations, and the G1/G2 offline gates RS-20260823T225753Z-0fbd48c3 | valid runoutcome: fail | tier: pilot | | The data factory, and the switch that was left off |
| Stage D0: fair relabelling of oracle-visited vs matched fair-D4 public states RS-20260823T205143Z-ead14c9d | valid runoutcome: fail | tier: pilot | | Perfect-information oracle |
| C0: paired D3 N7M6 minus D4 s5 contrast from retained per-game records RS-20260823T194200Z-42b113db | valid runoutcome: pass | tier: public-development | | Fair expectimax reference (D3/D4) |
| NNUE-class student on existing successor-closed D4 labels: within-root ordering probe RS-20260823T194142Z-946e3cd1 | valid runoutcome: fail | tier: pilot | | A learned survival estimate inside the reference search |
| Successor SCREEN: paired 256-game test of the reveal-construction leaf terms (two doses and a chain-to-crack bundle) against the unchanged fair D4 at seven strata with the memo engine, after the corpus value gate failed RS-20260823T131226Z-16564ed9 | valid runoutcome: fail | tier: public-development | | Chain-reveal leaf substrate |
| Seed-free corpus gate, then a paired 256-game SCREEN of fair D4 (seven strata, memo engine) with reveal-construction leaf terms against the unchanged leaf RS-20260823T110000Z-baecb816 | valid runoutcome: fail | tier: mechanics-only | | Chain-reveal leaf substrate |
| Paired SCREEN of fair D4 with the survival-instinct root filter (strict and literal) against unchanged fair D4 on 128 fresh development games RS-20260822T233343Z-12becce9 | valid runoutcome: inconclusive | tier: public-development | | Survival instinct — never place a disc where it can't clear vertically |
| CMA-ES over the eighteen fair-leaf weights inside the fast depth-4 five-stratum search, fresh training seeds per generation, one held-out paired screen RS-20260822T120736Z-662b39ca | valid runoutcome: fail | tier: public-development | | Evolving the leaf weights at the depth they are used |
| CHECK gates for the one-entry leaf memo: bit identity, search parity, determinism, interleaved timing on real probe seeds RS-20260822T074305Z-bd1697c8 | valid runoutcome: pass | tier: mechanics-only | | Remembering the last board |
| Prevalence, lead time and incremental lifetime signal of entombed numbered discs in the existing training corpus RS-20260822T051732Z-ceb257d7 | valid runoutcome: fail | tier: mechanics-only | | Entombed discs: the 3 that can never clear |
| Offline capacity and training sweep of the NNUE-shaped survival leaf on the existing corpus RS-20260822T024228Z-94090db1 | valid runoutcome: pass | tier: mechanics-only | | Is the small survival network as accurate as it can be? |
| Depth x chance-exactness factorial: the fifth ply at five and at seven strata, with an end-to-end reproduction control RS-20260821T205102Z-d89df4b5 | partial runoutcome: inconclusive | tier: public-development | | Fast engine |
| The two arms finding-09 left unfinished: reveal sampling on top of the fourth ply, and the depth-3 ladder at full joint coverage RS-20260821T192140Z-189fe392 | valid runoutcome: fail | tier: public-development | | Separate reveal and next-disc sampling |
| The two arms finding-09 left unfinished: reveal sampling on top of the fourth ply, and the depth-3 ladder at full joint coverage RS-20260821T181918Z-ea7076a3 | partial runoutcome: fail | tier: public-development | | Separate reveal and next-disc sampling |
| Depth x chance-exactness factorial: the fifth ply at five and at seven strata, with an end-to-end reproduction control RS-20260821T181917Z-9a34ba02 | partial runoutcome: fail | tier: public-development | | Fast engine |
| Stronger-teacher (D2 continuation) afterstate corpus: ranking gate and frozen override rule RS-20260821T134500Z-4b9d2f68 | valid runoutcome: fail | tier: pilot | | Distributional afterstate ranker |
| Scale-out stage 1: successor-closed fair-D4 search-value labels; held-out D4-ranking agreement RS-20260821T104500Z-77d21e90 | valid runoutcome: fail | tier: pilot | | Distributional afterstate ranker |
| Full training of the K=256 afterstate model, then the frozen top-two override rule on fresh origins RS-20260821T094500Z-1a7e3c55 | valid runoutcome: fail | tier: pilot | | Distributional afterstate ranker |
| Offline gate: calibrated top-two near-tie override of fair D4 by the frozen afterstate model RS-20260820T184500Z-63c0a8e2 | valid runoutcome: fail | tier: pilot | | Distributional afterstate ranker |
| Pilot iteration 3: K=256-scenario corpus, fresh held-out origins, full training RS-20260820T142500Z-8f4a2d17 | valid runoutcome: fail | tier: pilot | | Distributional afterstate ranker |
| Pilot iteration 2: K=64-scenario action-complete H40 corpus with fresh held-out origins RS-20260820T114500Z-2b7c9e31 | valid runoutcome: inconclusive | tier: pilot | | Distributional afterstate ranker |
| Pilot: action-complete H40 afterstate corpus and distributional ranker offline gate RS-20260820T094500Z-5c1e9a04 | valid runoutcome: inconclusive | tier: pilot | | Distributional afterstate ranker |