{"id":"arxiv/2605.01909","url":"https://arxiv.org/abs/2605.01909","tags":["stat.ML","cs.LG","math.ST","stat.ME"],"score":0,"title":"Extrapolation in Statistical Learning with Extreme Value Theory","domain":"stat.ML","source":"arxiv","authors":["Sebastian Engelke","Nicola Gnecco","Anne Sabourin"],"reviews":0,"subject":"stat.ML","summary":"Extreme value theory provides rigorous theory and statistical tools for extrapolation in machine learning, particularly in settings where traditional methods struggle due to data scarcity in the tails. A broad range of tasks benefit from these advances, including regression and classification beyond the training data, extreme quantile regression, supervised and unsupervised dimension reduction, generative artificial intelligence and anomaly detection. This review synthesizes recent developments in these fields at the intersection of statistical learning and extreme value theory, with a focus…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-05-03","reactions":{}}