{"id":"arxiv/2605.07908","url":"https://arxiv.org/abs/2605.07908","tags":["math.ST","cs.AI","cs.LG"],"score":0,"title":"Statistical inference with belief functions: A survey","domain":"math.ST","source":"arxiv","authors":["Fabio Cuzzolin"],"reviews":0,"subject":"math.ST","summary":"Belief functions are a powerful and popular framework for the mathematical characterisation of uncertainty, in particular in situations in which lack of data renders learning a probability distribution for the problem impractical. The first step in a reasoning chain based on belief functions is inference: how to learn a belief measure from the available data. In this survey we focus, in particular, on making inference from statistical data, and review the most significant contributions in the area.","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-08","reactions":{}}