A unified framework for supervised learning of semantic models

Yicheng Wen, Soumalya Sarkar, Asok Ray, Xin Jin, Thyagaraju Damarla

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

    Abstract

    Patterns of interest in dynamical systems are often represented by a number of semantic features such as probabilistic finite state automata (PFSA) and cross machines over possibly different alphabets. Previous publications have reported a Hilbert space formulation of PFSA over the same alphabet. This paper introduces an isomorphism between the Hilbert space of PFSA and the Euclidean space to improve the computational efficiency of algebraic operations. Furthermore, this formulation is extended to cross machines and it shows that these semantic features can be structured in a unified mathematical framework. In this framework, an algorithm of supervised learning is formulated for generating semantic features in the setting of linear discriminant analysis (LDA). The proposed algorithm has the flexibility for adaptation under different environments by tuning a set of parameters that can be updated autonomously or be specified by the human user. The proposed algorithm has been validated on real-life data for target detection as applied to border control.

    Original languageEnglish (US)
    Title of host publication2012 American Control Conference, ACC 2012
    Pages2183-2188
    Number of pages6
    StatePublished - 2012
    Event2012 American Control Conference, ACC 2012 - Montreal, QC, Canada
    Duration: Jun 27 2012Jun 29 2012

    Other

    Other2012 American Control Conference, ACC 2012
    CountryCanada
    CityMontreal, QC
    Period6/27/126/29/12

    Fingerprint

    Supervised learning
    Finite automata
    Semantics
    Hilbert spaces
    Discriminant analysis
    Computational efficiency
    Target tracking
    Dynamical systems
    Tuning

    All Science Journal Classification (ASJC) codes

    • Electrical and Electronic Engineering

    Cite this

    Wen, Y., Sarkar, S., Ray, A., Jin, X., & Damarla, T. (2012). A unified framework for supervised learning of semantic models. In 2012 American Control Conference, ACC 2012 (pp. 2183-2188). [6314719]
    Wen, Yicheng ; Sarkar, Soumalya ; Ray, Asok ; Jin, Xin ; Damarla, Thyagaraju. / A unified framework for supervised learning of semantic models. 2012 American Control Conference, ACC 2012. 2012. pp. 2183-2188
    @inproceedings{1a3e65f16eb9495b9ec9bd1bb44a41ad,
    title = "A unified framework for supervised learning of semantic models",
    abstract = "Patterns of interest in dynamical systems are often represented by a number of semantic features such as probabilistic finite state automata (PFSA) and cross machines over possibly different alphabets. Previous publications have reported a Hilbert space formulation of PFSA over the same alphabet. This paper introduces an isomorphism between the Hilbert space of PFSA and the Euclidean space to improve the computational efficiency of algebraic operations. Furthermore, this formulation is extended to cross machines and it shows that these semantic features can be structured in a unified mathematical framework. In this framework, an algorithm of supervised learning is formulated for generating semantic features in the setting of linear discriminant analysis (LDA). The proposed algorithm has the flexibility for adaptation under different environments by tuning a set of parameters that can be updated autonomously or be specified by the human user. The proposed algorithm has been validated on real-life data for target detection as applied to border control.",
    author = "Yicheng Wen and Soumalya Sarkar and Asok Ray and Xin Jin and Thyagaraju Damarla",
    year = "2012",
    language = "English (US)",
    isbn = "9781457710957",
    pages = "2183--2188",
    booktitle = "2012 American Control Conference, ACC 2012",

    }

    Wen, Y, Sarkar, S, Ray, A, Jin, X & Damarla, T 2012, A unified framework for supervised learning of semantic models. in 2012 American Control Conference, ACC 2012., 6314719, pp. 2183-2188, 2012 American Control Conference, ACC 2012, Montreal, QC, Canada, 6/27/12.

    A unified framework for supervised learning of semantic models. / Wen, Yicheng; Sarkar, Soumalya; Ray, Asok; Jin, Xin; Damarla, Thyagaraju.

    2012 American Control Conference, ACC 2012. 2012. p. 2183-2188 6314719.

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

    TY - GEN

    T1 - A unified framework for supervised learning of semantic models

    AU - Wen, Yicheng

    AU - Sarkar, Soumalya

    AU - Ray, Asok

    AU - Jin, Xin

    AU - Damarla, Thyagaraju

    PY - 2012

    Y1 - 2012

    N2 - Patterns of interest in dynamical systems are often represented by a number of semantic features such as probabilistic finite state automata (PFSA) and cross machines over possibly different alphabets. Previous publications have reported a Hilbert space formulation of PFSA over the same alphabet. This paper introduces an isomorphism between the Hilbert space of PFSA and the Euclidean space to improve the computational efficiency of algebraic operations. Furthermore, this formulation is extended to cross machines and it shows that these semantic features can be structured in a unified mathematical framework. In this framework, an algorithm of supervised learning is formulated for generating semantic features in the setting of linear discriminant analysis (LDA). The proposed algorithm has the flexibility for adaptation under different environments by tuning a set of parameters that can be updated autonomously or be specified by the human user. The proposed algorithm has been validated on real-life data for target detection as applied to border control.

    AB - Patterns of interest in dynamical systems are often represented by a number of semantic features such as probabilistic finite state automata (PFSA) and cross machines over possibly different alphabets. Previous publications have reported a Hilbert space formulation of PFSA over the same alphabet. This paper introduces an isomorphism between the Hilbert space of PFSA and the Euclidean space to improve the computational efficiency of algebraic operations. Furthermore, this formulation is extended to cross machines and it shows that these semantic features can be structured in a unified mathematical framework. In this framework, an algorithm of supervised learning is formulated for generating semantic features in the setting of linear discriminant analysis (LDA). The proposed algorithm has the flexibility for adaptation under different environments by tuning a set of parameters that can be updated autonomously or be specified by the human user. The proposed algorithm has been validated on real-life data for target detection as applied to border control.

    UR - http://www.scopus.com/inward/record.url?scp=84869465804&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84869465804&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:84869465804

    SN - 9781457710957

    SP - 2183

    EP - 2188

    BT - 2012 American Control Conference, ACC 2012

    ER -

    Wen Y, Sarkar S, Ray A, Jin X, Damarla T. A unified framework for supervised learning of semantic models. In 2012 American Control Conference, ACC 2012. 2012. p. 2183-2188. 6314719