Add regularization parameter to pair selection weighting#99
Merged
kronosapiens merged 1 commit intomainfrom Mar 13, 2026
Merged
Add regularization parameter to pair selection weighting#99kronosapiens merged 1 commit intomainfrom
kronosapiens merged 1 commit intomainfrom
Conversation
Introduces optional r parameter (0–1) to blend pair selection weights toward uniform. At r=1 (default), full coverage×proximity×position weighting applies. At r=0, all pairs equally likely. At r=0.5, weights are interpolated 50/50, fixing the issue where under-observed entries get starved while already-ranked items dominate pair selection. Fixes skew caused by earlier bugs where some entries received disproportionate comparisons. New judges now see a more balanced mix of entries instead of only low-ranked ones, while still biasing toward coverage and uncertainty reduction. Co-Authored-By: Claude Haiku 4.5 <[email protected]>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Introduces optional
rparameter (0–1) toactiveSelect()for blending pair selection weights toward uniform. At r=1 (default), full coverage×proximity×position weighting applies; at r=0, all pairs are equally likely; at r=0.5, weights interpolate 50/50.This fixes the skew where under-observed entries get starved while already-highly-ranked entries dominate pair selection—particularly problematic when earlier bugs caused some entries (like athenor) to receive disproportionate comparisons. New judges now see a more balanced mix of entries instead of only low-ranked ones, while still biasing toward coverage and uncertainty reduction.
Sets production value to r=0.5 to smoothly regularize weighting without losing signal. Includes two new tests validating r behavior (uniform at r=0, compressed spread at r<1).