U-Statistics

Latest papers

11 papers2 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Pairwise AUC Optimization Needs Corrective Power: A Unified View
Oct 15, 2026Jia Chen, Zhiyong Yang, Shilong Bao +2U-Statistics Learning to Rank
School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, China · Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
Ascent Conditions for Length-Normalized Single-Rollout Policy Updates
Oct 10, 2026Levin David SchwabU-Statistics Policy Gradient Methods
When Pair Count Is Not the Sample Size: What All-Pairs Agent Comparisons Estimate
Sep 26, 2026Wei-Jung HuangU-Statistics Pairwise Comparison
Independent Researcher
INCLAIR: Inception-Based Longitudinal Clinical Anomaly Detection with Informed Reasoning
Jul 29, 2026Maxx Richard Rahman, Wolfgang MaassU-Statistics Interpretable Anomaly Detection
German Research Center for Artificial Intelligence (DFKI), Germany · Saarland University, Germany
On Finite-sample Concentration of Median of Incomplete U-Statistics
May 30, 2026Nong Minh Hieu, Antoine LedentRobust Estimation U-Statistics
Singapore Management University, School of Computing and Information Systems
Learning U-Statistics with Active Inference
May 12, 2026Xiaoning Wang, Yuyang Huo, Liuhua Peng +1Active Inference U-Statistics
School of Statistics and Data Sciences, LPMC, KLM-DASR and LEBPS, Nankai University, Tianjin, China · School of Mathematics and Statistics, The University of Melbourne, Melbourne, Australia
Generalized Distributional Alignment Games for Unbiased Answer-Level Fine-Tuning
May 4, 2026Mehryar Mohri, Jon Schneider, Yutao ZhongFine-Tuning LLM Alignment
Google Research & CIMS · New York, NY 10011 · Google Research
On Gossip Algorithms for Machine Learning with Pairwise Objectives
Mar 25, 2026Igor Colin, Aurélien Bellet, Stephan Clémençon +1U-Statistics Decentralized Optimization
LTCI, S2A, IP Paris · LTCI, Télécom Paris, Institut Polytechnique de Paris · PREMEDICAL +4
Accurate, private, secure, federated U-statistics with higher degree
Mar 2, 2026Quentin Sinh, Jan RamonU-Statistics Differentially Private Federated Learning
MAGNET · MAGNET, Univ. Lille, INRIA, CNRS, UMR 9189 - CRIStAL, F-59000 Lille, France.
Demystifying Group Relative Policy Optimization: Its Policy Gradient is a U-Statistic
Mar 1, 2026Hongyi Zhou, Kai Ye, Erhan Xu +4U-Statistics Group Relative Policy Optimization
Department of Mathematics, Tsinghua University · Department of Statistics, London School of Economics and Political Science · School of Mathematics, University of Birmingham +2
Maximum Mean Discrepancy with Unequal Sample Sizes via Generalized U-Statistics
Dec 16, 2025Aaron Wei, Milad Jalali, Danica J. SutherlandTwo-Sample Testing Maximum Mean Discrepancy
University of British Columbia · Independent researcher, Vancouver, Canada · Alberta Machine Intelligence Institute

Oct 15, 2026J/K move · Enter open · S save

Pairwise AUC Optimization Needs Corrective Power: A Unified View

Jia Chen, Zhiyong Yang, Shilong Bao, Qianqian Xu+1

School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, China · Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China

U-Statistics Learning to Rank Stochastic Gradient Descent

Open paper AskPDF arXiv

Direct pairwise AUC-surrogate training is attractive because it trains toward the ranking metric used at test time, especially under class imbalance. This creates a direct alignment between the training objective and the evaluation metric, but such metric alignment alone does not guarantee that stochastic training remains corrective. We study this gap through Corrective Power: whether pairwise mini-batch updates produce stronger and more reliable improvements for more severe positive-negative ranking errors. Our main observation is that direct pairwise training can enter regions where severe errors remain common but no longer produce useful corrective motion. In such regions, even the mini-batch noise that would normally help SGD move along the steepest severe-error correction direction can become much weaker. In this paper, we seek a theoretical explanation for this by exploring the degenerate property of the U-statistics formed from the pairwise gradients therein. Our results provide a mechanistic explanation for several empirical patterns: (a) squared-hinge loss is often an effective AUC surrogate; (b) cross-entropy warm-up can help models that enter the AUC phase from a poor starting region; and (c) parameter-efficient fine-tuning from a pretrained model can reduce that need by starting closer to a favorable ranking geometry. Experiments on 8 CV tasks and 8 NLP tasks, using ResNet50, DenseNet121, DistilBERT-base, and Qwen2.5-1.5B-Instruct, support these predictions.