{"id":"arxiv/2611.18480","url":"https://arxiv.org/abs/2611.18480","tags":["stat.ML","cs.LG","math.OC"],"score":0,"title":"Bregman geometry in regular exponential families: dual representations and relative smoothness","domain":"stat.ML","source":"arxiv","authors":["Mame Diarra Fall","Éric Barat","Nicolas Dobigeon"],"reviews":0,"subject":"stat.ML","summary":"Regular exponential-family distributions are ubiquitous in statistics due to their appealing and convenient properties. Notably, they can be expressed in terms of Bregman divergences, revealing fundamental connections between statistics, convex analysis, and information geometry. However, this geometric structure has not been systematically exploited to characterize these distributions with respect to their expectation parametrization. Building on a projective characterization, we establish a general construction of equivalent dual Bregman representations, together with normalization and…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-11-13","reactions":{}}