DocumentCode
2022867
Title
The Information Bottleneck Revisited or How to Choose a Good Distortion Measure
Author
Harremoes, P. ; Tishby, N.
Author_Institution
Centrum voor Wiskunde en Inf., Amsterdam
fYear
2007
fDate
24-29 June 2007
Firstpage
566
Lastpage
570
Abstract
It is well-known that the information bottleneck method and rate distortion theory are related. Here it is described how the information bottleneck can be considered as rate distortion theory for a family of probability measures where information divergence is used as distortion measure. It is shown that the information bottleneck method has some properties that are not shared with rate distortion theory based on any other divergence measure. In this sense the information bottleneck method is unique.
Keywords
distortion; information theory; distortion measure; information bottleneck; information divergence; probability measures; rate distortion theory; Computer science; Distortion measurement; Information retrieval; Kernel; Loss measurement; Random variables; Rate distortion theory; Rate-distortion; Time measurement;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory, 2007. ISIT 2007. IEEE International Symposium on
Conference_Location
Nice
Print_ISBN
978-1-4244-1397-3
Type
conf
DOI
10.1109/ISIT.2007.4557285
Filename
4557285
Link To Document