LambdaRankIC derives closed-form lambda gradients for pairwise rank swaps to directly optimize Rank IC within the LambdaRank framework, outperforming regression and NDCG losses on simulated and real financial data.
Journal of the Royal Statistical Society
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LambdaRankIC: Directly Optimizing Rank IC for Financial Prediction
LambdaRankIC derives closed-form lambda gradients for pairwise rank swaps to directly optimize Rank IC within the LambdaRank framework, outperforming regression and NDCG losses on simulated and real financial data.