jcreek/CosmicClash
1
0
Fork 0
mirror of https://github.com/jcreek/CosmicClash.git synced 2026-09-16 16:32:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

test(training): require multi-seed curriculum evaluation

Browse Source
This commit is contained in:
Josh Creek
2026-09-01 17:26:44 +01:00
parent 533ac1afab
commit 9004800326
4 changed files with 32 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
TRAINING.md
+5 -1
View File
@@ -855,7 +855,11 @@ real tail, the same way this one now has been.
Stage 6's `league` opponent mode samples a historical exported policy at each
episode reset. Each later stage preserves the preceding shaping and adds one
new difficulty.
new difficulty. The generation-5 orchestrator evaluates every candidate
against every reference on three independent paired seeds (`1, 19, 43`) before
advancing; pass `--evaluation-seeds` only when deliberately running a
different, recorded experiment. This avoids promoting a policy from a single
side-biased starting-state sequence.
The physical-side gate is separate from the model-vs-model score. A paired
side swap can make an identical policy appear perfectly balanced overall even