The Power of Amnesia

The Power of Amnesia (1994)

Cached

Download Links

[www.cs.huji.ac.il]

Other Repositories/Bibliography

CiteULike
DBLP

by Dana Ron , Yoram Singer , Naftali Tishby

Venue:	Machine Learning
Citations:	69 - 4 self

Versions

Version History

Metadata Version 3

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	The Power of Amnesia	user correction - Legacy Corrections
AUTHOR NAME	Dana Ron	user correction - Legacy Corrections
AUTHOR AFFIL	Institute of Computer Science and; Center for Neural Computation	user correction - Legacy Corrections
AUTHOR ADDR	Hebrew University, Jerusalem 91904, Israel	user correction - Legacy Corrections
AUTHOR NAME	Yoram Singer	user correction - Legacy Corrections
AUTHOR AFFIL	Institute of Computer Science and; Center for Neural Computation	user correction - Legacy Corrections
AUTHOR ADDR	Hebrew University, Jerusalem 91904, Israel	user correction - Legacy Corrections
AUTHOR NAME	Naftali Tishby	user correction - Legacy Corrections
AUTHOR AFFIL	Institute of Computer Science and; Center for Neural Computation	user correction - Legacy Corrections
AUTHOR ADDR	Hebrew University, Jerusalem 91904, Israel	user correction - Legacy Corrections
ABSTRACT	We propose a learning algorithm for a variable memory length Markov process. Human communication, whether given as text, handwriting, or speech, has multi characteristic time scales. On short scales it is characterized mostly by the dynamics that generate the process, whereas on large scales, more syntactic and semantic information is carried. For that reason the conventionally used fixed memory Markov models cannot capture effectively the complexity of such structures. On the other hand using long memory models uniformly is not practical even for as short memory as four. The algorithm we propose is based on minimizing the statistical prediction error by extending the memory, or state length, adaptively, until the total prediction error is sufficiently small. We demonstrate the algorithm by learning the structure of natural English text and applying the learned model to the correction of corrupted text. Using less than 3000 states the model's performance is far superior to that of fixe...	user correction - Legacy Corrections
YEAR	1994	INFERENCE
VENUE	Machine Learning	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
PAGES	176--183	INFERENCE
VOLUME	6	INFERENCE
CITATIONS	8 found	ParsCit 1.0