From 6e90f1145b2c5c0f2dcc82d21ec45432274b8cad Mon Sep 17 00:00:00 2001 From: tslil Date: Fri, 11 Sep 2026 12:56:12 +0100 Subject: move from single candidate td-leaf with gating to tournament of a population The goal in this change was to overcome the rock-paper-scissors phenomenon that was happening in capability space by having on a single candidate learn to beat the most recent version of itself. In some sense this lead to rotating around in capability space instead of chasing monotonic improvements. I hope that flipping to a system in which many (N=5) candidates play each other and learn from their own set of diverse games in this pool, combined with --replay-iters, will lead to a given network having to have a much more comprehensive understanding of the space of tactics. I haven't yet had time to train this thorougly, will run those experiments on my desktop computer. --- .gitignore | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to '.gitignore') diff --git a/.gitignore b/.gitignore index b013cd7..0d00588 100644 --- a/.gitignore +++ b/.gitignore @@ -1,15 +1,12 @@ *.o data/ -include/cand_weights_*.c +include/cand_* ct1986* ctlm* pptdb* cttei* geminict* selfplay* -match* __pycache__/ -production-*.log buildroot* -test/ .cache -- cgit v1.3.1