Probabilistic Classification: Venn Predictors

Vladimir Vovk, Alexander Gammerman and Glenn Shafer
Additional contact information
Vladimir Vovk: University of London, Royal Holloway
Alexander Gammerman: University of London, Royal Holloway
Glenn Shafer: Rutgers University

Chapter Chapter 6 in Algorithmic Learning in a Random World, 2022, pp 157-179 from Springer

Abstract: Abstract In this chapter we discuss positive results about probabilistic prediction in the case of classification. For that, we introduce a new kind of predictors, Venn predictors, which produce predictions satisfying a natural property of validity at the price of lowering the bar as compared with the previous chapter. To gain some wiggle room, Venn predictors are allowed to announce several probability distributions as their prediction, and the overall prediction is considered valid whenever any of its components is valid. Such a multiprobability prediction is useful if its components are close to each other, in which case we have an approximate probabilistic prediction; if not, we may be in a situation of Knightean uncertainty, and the multiprobability prediction may be considered as a useful warning.

Keywords: Classification; Venn prediction; Venn–Abers predictors (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-031-06649-8_6

Ordering information: This item can be ordered from
http://www.springer.com/9783031066498

DOI: 10.1007/978-3-031-06649-8_6

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().