Sökning: id:"swepub:oai:DiVA.org:liu-13105" >
Concentrated or non...
Concentrated or non-concentrated discrete distributions are almost independent
-
- Ekdahl, Magnus (författare)
- Linköpings universitet,Matematisk statistik,Tekniska högskolan
-
- Koski, Timo (författare)
- Linköpings universitet,Matematisk statistik,Tekniska högskolan
-
- Ohlson, Martin (författare)
- Linköpings universitet,Matematisk statistik,Tekniska högskolan
-
(creator_code:org_t)
- 2007
- Engelska.
- Relaterad länk:
-
https://urn.kb.se/re...
Abstract
Ämnesord
Stäng
- The task of approximating a simultaneous distribution with a product of distributions in a single variable is important in the theory and applications of classification and learning, probabilistic reasoning, and random algmithms. The evaluation of the goodness of this approximation by statistical independence amounts to bounding uniformly upwards the difference between a joint distribution and the product of the distributions (marginals). In this paper we develop a bound that uses information about the most probable state to find a sharp estimate, which is often as sharp as possible. We also examine the extreme cases of concentration and non-conccntmtion, respectively, of the approximated distribution.
Ämnesord
- NATURVETENSKAP -- Matematik (hsv//swe)
- NATURAL SCIENCES -- Mathematics (hsv//eng)
Nyckelord
- MATHEMATICS
- MATEMATIK
Publikations- och innehållstyp
- vet (ämneskategori)
- ovr (ämneskategori)