Approximate maximum likelihood source separation using the natural gradient

Seungjin Choi, A. Cichocki, Liqing Zhang, S. Amari

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Citations (Scopus)

Abstract

This paper addresses a maximum likelihood approach to source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We present an objective function that is an approximate likelihood function based on the Laplace approximation. Then we derive a natural gradient adaptation algorithm which maximizes the corresponding approximate likelihood function. Useful behavior of the proposed method is verified by numerical experiments.

Original languageEnglish
Title of host publication2001 IEEE 3rd Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2001
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages235-238
Number of pages4
ISBN (Electronic)0780367200
DOIs
Publication statusPublished - 2001
Externally publishedYes
Event3rd IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2001 - Taoyuan, Taiwan, Province of China
Duration: 20 Mar 200123 Mar 2001

Publication series

NameIEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC
Volume2001-January

Conference

Conference3rd IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2001
Country/TerritoryTaiwan, Province of China
CityTaoyuan
Period20/03/0123/03/01

Keywords

  • Additive white noise
  • Artificial intelligence
  • Computer science
  • Context
  • Information systems
  • Maximum likelihood estimation
  • Signal processing
  • Signal processing algorithms
  • Source separation
  • Statistical analysis

Fingerprint

Dive into the research topics of 'Approximate maximum likelihood source separation using the natural gradient'. Together they form a unique fingerprint.

Cite this