Pattern Classification in symbolic streams via semantic annihilation of information

Ishanu Chattopadhyay, Yicheng Wen, Asok Ray

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

1 Citation (Scopus)

Abstract

We propose a technique for pattern identification in symbolic streams via selective erasure of observed symbols, in cases where the patterns of interest are represented as Probabilistic Finite State Automata (PFSA). We define an additive abelian group for a slightly restricted set of probabilistic machines, and the group sum is used to formulate pattern-specific semantic annihilators. The annihilators attempt to identify pre-specified patterns via removal of inter-symbol correlations from observed sequences, thereby turning them into symbolic white noise. Thus a perfect annihilation corresponds to a pattern match. This approach of classification via information annihilation is shown to be strictly advantageous, with theoretical guarantees, for a large class of PFSA models. The results are supported by simulation experiments.

Original languageEnglish (US)
Title of host publicationProceedings of the 2010 American Control Conference, ACC 2010
Pages492-497
Number of pages6
StatePublished - 2010
Event2010 American Control Conference, ACC 2010 - Baltimore, MD, United States
Duration: Jun 30 2010Jul 2 2010

Other

Other2010 American Control Conference, ACC 2010
CountryUnited States
CityBaltimore, MD
Period6/30/107/2/10

Fingerprint

Finite automata
Pattern recognition
Semantics
White noise
Experiments

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering

Cite this

Chattopadhyay, I., Wen, Y., & Ray, A. (2010). Pattern Classification in symbolic streams via semantic annihilation of information. In Proceedings of the 2010 American Control Conference, ACC 2010 (pp. 492-497). [5530857]
Chattopadhyay, Ishanu ; Wen, Yicheng ; Ray, Asok. / Pattern Classification in symbolic streams via semantic annihilation of information. Proceedings of the 2010 American Control Conference, ACC 2010. 2010. pp. 492-497
@inproceedings{f829ddec11b941f48e941e55b8f833c4,
title = "Pattern Classification in symbolic streams via semantic annihilation of information",
abstract = "We propose a technique for pattern identification in symbolic streams via selective erasure of observed symbols, in cases where the patterns of interest are represented as Probabilistic Finite State Automata (PFSA). We define an additive abelian group for a slightly restricted set of probabilistic machines, and the group sum is used to formulate pattern-specific semantic annihilators. The annihilators attempt to identify pre-specified patterns via removal of inter-symbol correlations from observed sequences, thereby turning them into symbolic white noise. Thus a perfect annihilation corresponds to a pattern match. This approach of classification via information annihilation is shown to be strictly advantageous, with theoretical guarantees, for a large class of PFSA models. The results are supported by simulation experiments.",
author = "Ishanu Chattopadhyay and Yicheng Wen and Asok Ray",
year = "2010",
language = "English (US)",
isbn = "9781424474264",
pages = "492--497",
booktitle = "Proceedings of the 2010 American Control Conference, ACC 2010",

}

Chattopadhyay, I, Wen, Y & Ray, A 2010, Pattern Classification in symbolic streams via semantic annihilation of information. in Proceedings of the 2010 American Control Conference, ACC 2010., 5530857, pp. 492-497, 2010 American Control Conference, ACC 2010, Baltimore, MD, United States, 6/30/10.

Pattern Classification in symbolic streams via semantic annihilation of information. / Chattopadhyay, Ishanu; Wen, Yicheng; Ray, Asok.

Proceedings of the 2010 American Control Conference, ACC 2010. 2010. p. 492-497 5530857.

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

TY - GEN

T1 - Pattern Classification in symbolic streams via semantic annihilation of information

AU - Chattopadhyay, Ishanu

AU - Wen, Yicheng

AU - Ray, Asok

PY - 2010

Y1 - 2010

N2 - We propose a technique for pattern identification in symbolic streams via selective erasure of observed symbols, in cases where the patterns of interest are represented as Probabilistic Finite State Automata (PFSA). We define an additive abelian group for a slightly restricted set of probabilistic machines, and the group sum is used to formulate pattern-specific semantic annihilators. The annihilators attempt to identify pre-specified patterns via removal of inter-symbol correlations from observed sequences, thereby turning them into symbolic white noise. Thus a perfect annihilation corresponds to a pattern match. This approach of classification via information annihilation is shown to be strictly advantageous, with theoretical guarantees, for a large class of PFSA models. The results are supported by simulation experiments.

AB - We propose a technique for pattern identification in symbolic streams via selective erasure of observed symbols, in cases where the patterns of interest are represented as Probabilistic Finite State Automata (PFSA). We define an additive abelian group for a slightly restricted set of probabilistic machines, and the group sum is used to formulate pattern-specific semantic annihilators. The annihilators attempt to identify pre-specified patterns via removal of inter-symbol correlations from observed sequences, thereby turning them into symbolic white noise. Thus a perfect annihilation corresponds to a pattern match. This approach of classification via information annihilation is shown to be strictly advantageous, with theoretical guarantees, for a large class of PFSA models. The results are supported by simulation experiments.

UR - http://www.scopus.com/inward/record.url?scp=77957796119&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957796119&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781424474264

SP - 492

EP - 497

BT - Proceedings of the 2010 American Control Conference, ACC 2010

ER -

Chattopadhyay I, Wen Y, Ray A. Pattern Classification in symbolic streams via semantic annihilation of information. In Proceedings of the 2010 American Control Conference, ACC 2010. 2010. p. 492-497. 5530857