Generalize higher-order moments in independent component analysis

Author

Coleman, Jeffrey O.

Author_Institution

US Naval Res. Lab., Washington, DC, USA

Volume

1

fYear

2000

fDate

2000

Firstpage

153

Abstract

In independent component analysis (ICA), random-variable independence is often equated with factorization of the joint moments, expectations of products of powers. This paper shows that many nonpower functions are equally useful: if E[f(X)g(Y)] factors into E[f(X)]E[g(Y)] for every f and g from an independence class, then random variables X and Y are independent. Examples of and sufficient conditions for independence classes are presented for bounded random variables

Keywords

identification; random processes; set theory; signal processing; statistical analysis; Borel functions; Borel sets; bounded random variables; generalized higher-order moments; independence classes; independent component analysis; independent random variables; joint moments factorization; nonpower functions; products of powers expectations; random-variable independence; signal procesing; sufficient conditions; sufficient identification criteria; Bismuth; Independent component analysis; Random variables; Sufficient conditions;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on

Conference_Location

Istanbul

ISSN

1520-6149

Print_ISBN

0-7803-6293-4

Type

conf

DOI

10.1109/ICASSP.2000.861896

Filename

861896