Skip to content

KEYRING

Build a board from scratch, ranking each next piece by three signals learned from past strong boards. Reached 460 in a board family no earlier search had cracked.

by Raphaël Anjou

experimentsolvermeasurednew basin460/480matched edgesBuild boards upLearned guidanceUpdated 2026-07-21
Reproducestochastic — won't reproduce exactly; the board is verifiablere-verifies a stored boardBudget: not logged (exploratory run, not the standardized single-core bench)
Pipeline
  1. 1
    beam producernot yet written up

    Build a board from empty, ranking each next piece by three learned signals voted together

    carries: Three corpus signals: a 2x2-patch log-odds prior, piece-in-position and piece-pair frequencies

  2. 2
    refinementnot yet written up

    A local-refinement tail lifts the high-450s construction to the committed 460

Complexity
Time
beam search O(W·n·|moves|) for beam width W over n = 256 placements; prior build O(B) in the corpus
Space
O(H) for the patch table (H distinct 2×2 patches seen) + O(W·n) live beams

The learned priors are a one-off preprocessing pass; the run cost is the beam search they bias.

Hardware & run
Native runCPU only
Cores
8
RAM
16 GiB
GPUs
0
CPU
Apple M1
Machine
MacBook (Apple M1, 8 cores)
Budget
not logged (exploratory run, not the standardized single-core bench)
Start
stochastic; see the run's repro command
Verify the stored board

stochastic — won't reproduce exactly; the board is verifiable

This verifies the board, not the search that produced it.

just research-record-boards
Code & data on GitHub

Keep exploring

Referenced by

Page sourceView as Markdown