On the learnability and usage of acyclic probabilistic finite automata

Dana Ron, Yoram Singer, Naftali Tishby

Research output: Chapter in Book/Report/Conference proceedingConference contribution

49 Scopus citations

Abstract

We propose and analyze a distribution learning algorithm for a subclass of Acyclic Probabilistic Finite Automata (APFA). This subclass is characterized by a certain distinguishability property of the automata's states. Though hardness results are known for learning distributions generated by general APFAs, we prove that our algorithm can indeed efficiently learn distributions generated by the subclass of APFAs we consider. In particular, we show that the KL-divergence between the distribution generated by the target source and the distribution generated by our hypothesis can be made small with high confidence in polynomial time. We present two applications of our algorithm. In the first, we show how to model cursively written letters. The resulting models are part of a complete cursive handwriting recognition system. In the second application we demonstrate how APFAs can be used to build multiple-pronunciation models for spoken words. We evaluate the APFA based pronunciation models on labeled speech data. The good performance (in terms of the log-likelihood obtained on test data) achieved by the APFAs and the incredibly small amount of time needed for learning suggests that the learning algorithm of AP-FAs might be a powerful alternative to commonly used probabilistic models.

Original languageEnglish (US)
Title of host publicationProceedings of the 8th Annual Conference on Computational Learning Theory, COLT 1995
PublisherAssociation for Computing Machinery, Inc
Pages31-40
Number of pages10
ISBN (Electronic)0897917235, 9780897917230
DOIs
StatePublished - Jul 5 1995
Event8th Annual Conference on Computational Learning Theory, COLT 1995 - Santa Cruz, United States
Duration: Jul 5 1995Jul 8 1995

Publication series

NameProceedings of the 8th Annual Conference on Computational Learning Theory, COLT 1995
Volume1995-January

Other

Other8th Annual Conference on Computational Learning Theory, COLT 1995
CountryUnited States
CitySanta Cruz
Period7/5/957/8/95

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Artificial Intelligence
  • Software

Fingerprint Dive into the research topics of 'On the learnability and usage of acyclic probabilistic finite automata'. Together they form a unique fingerprint.

Cite this