Robust Bayesian mixture modeling
US7636651B2 · kind B2 · utility
4Cited by
8References
64Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Nov 28, 2003 |
| Grant date | Dec 22, 2009 |
| Priority date | — |
| Expiry date | Apr 3, 2025 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N7/01
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A Bayesian treatment of mixture models is based on individual components having Student distributions, which have heavier tails compared to the exponentially decaying tails of Gaussians. The mixture of Student distribution components is characterized by a set of modeling parameters. Tractable approximations of the posterior distributions of individual modeling parameters are optimized and used to generate a data model for a set of input data.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.