Stackruns

Run T-0164 rejected

· Evidence: SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.

Hypothesis

Initializing the accumulator with the input bias rather than zero, so network output heads across standard, AVX2, AVX-512, and single-layer implementations no longer need to add it, will leave the search tree identical while increasing engine speed.

Wording edited for publication from the German entry; the register carries the author’s own text.

Stages

Elo against the predecessorElo intervals: short stage +8.63 ± 4.56; long stage −0.44 ± 2.07.0+5+10+15short stage+8.63 ± 4.56short stage · passed+8.63 ± 4.56 Elo · 4,066 gameslong stage−0.44 ± 2.07long stage · rejected−0.44 ± 2.07 Elo · 20,386 games
Point estimate with the reported 95-percent half-width, per stage. The dashed line is zero. Estimates from a stopped sequential test are biased by the stopping rule; the interval is nominal.
Field short stagelong stage
Verdictpassedrejected
Elo Strength difference to the opponent, estimated from the games, with a 95-percent half-width where the oracle reported one. Never an absolute rating.+8.63 ± 4.56−0.44 ± 2.07
nElo Normalised Elo: the Elo difference divided by the spread of the game results, so that the SPRT bounds mean the same at different time controls.+20.23 ± 10.68−1.02 ± 4.77
Games4,06620,386
Wins / draws / losses1,017 / 2,133 / 9164,897 / 10,566 / 4,923
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.5 / 304 / 1312 / 409 / 317 / 1865 / 6447 / 1855 / 9
LLR Log-likelihood ratio, the running evidence of an SPRT. It starts at 0; a stage passes at the upper bound and fails at the lower one.+2.97−2.97
Bounds−2.94 … +2.94−2.94 … +2.94
SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.[0.0, 5.0][0.0, 5.0]
Error rates α / β0.05 / 0.050.05 / 0.05
Modelnormalizednormalized
time control Base time plus increment per move in seconds, e.g. 8+0.08: eight seconds per game plus 0.08 seconds per move.8+0.0840+0.4
BookAB
Opponentvs predecessorvs predecessor
Duration30 min13 h 17 min

Provenance

Date
2026-10-06
Started
2026-10-06 11:26:39 UTC
Duration
13 h 48 min
Author
Agent X
Epoch
7
Gatekeeper
passed
Bench nodes
851