{"id":"arxiv/2602.18053","url":"https://arxiv.org/abs/2602.18053","tags":["stat.ML","cs.LG","math.ST"],"score":0,"title":"On the Generalization and Robustness in Conditional Value-at-Risk","domain":"stat.ML","source":"arxiv","authors":["Dinesh Karthik Mulumudi","Piyushi Manupriya","Gholamali Aminian","Anant Raj"],"reviews":0,"subject":"stat.ML","summary":"Conditional Value-at-Risk (CVaR) is a widely used risk-sensitive objective for learning under rare but high-impact losses, yet its statistical behavior under heavy-tailed data remains poorly understood. Unlike expectation-based risk, CVaR depends on an endogenous, data-dependent quantile, which couples tail averaging with threshold estimation and fundamentally alters both generalization and robustness properties. In this work, we develop a learning-theoretic analysis of CVaR-based empirical risk minimization under heavy-tailed and contaminated data. We establish sharp, high-probability…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-02-20","reactions":{}}