Fast estimation of Hidden Markov Models via alpha-EM algorithm

Yasuo Matsuyama, Ryunosuke Hayashi, Ryota Yokote

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

    Abstract

    Fast estimation algorithms of Hidden Markov Models (HMMs), or alpha-HMMs, are presented. Such novel algorithms inherit speedup properties of the alpha-EM algorithm. Since the alpha-EM algorithm includes the traditional log-EM algorithm as its special case, the alpha-HMM also includes the traditional log-HMM as its special case. This generalization appears as the utilization of the past information which is the main device of the speedup. Since the memorization of the past information requires only little increase of computational load and memory, the iteration speedup directly appears as that of CPU time. Experimental results are given.

    Original languageEnglish
    Title of host publicationIEEE Workshop on Statistical Signal Processing Proceedings
    Pages89-92
    Number of pages4
    DOIs
    Publication statusPublished - 2011
    Event2011 IEEE Statistical Signal Processing Workshop, SSP 2011 - Nice
    Duration: 2011 Jun 282011 Jun 30

    Other

    Other2011 IEEE Statistical Signal Processing Workshop, SSP 2011
    CityNice
    Period11/6/2811/6/30

    Fingerprint

    Hidden Markov models
    EM Algorithm
    Markov Model
    Speedup
    CPU Time
    Estimation Algorithms
    Fast Algorithm
    Program processors
    Iteration
    Experimental Results
    Data storage equipment

    Keywords

    • alpha-EM algorithm
    • alpha-HMM
    • past information
    • speedup

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Applied Mathematics
    • Signal Processing
    • Computer Science Applications

    Cite this

    Matsuyama, Y., Hayashi, R., & Yokote, R. (2011). Fast estimation of Hidden Markov Models via alpha-EM algorithm. In IEEE Workshop on Statistical Signal Processing Proceedings (pp. 89-92). [5967835] https://doi.org/10.1109/SSP.2011.5967835

    Fast estimation of Hidden Markov Models via alpha-EM algorithm. / Matsuyama, Yasuo; Hayashi, Ryunosuke; Yokote, Ryota.

    IEEE Workshop on Statistical Signal Processing Proceedings. 2011. p. 89-92 5967835.

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

    Matsuyama, Y, Hayashi, R & Yokote, R 2011, Fast estimation of Hidden Markov Models via alpha-EM algorithm. in IEEE Workshop on Statistical Signal Processing Proceedings., 5967835, pp. 89-92, 2011 IEEE Statistical Signal Processing Workshop, SSP 2011, Nice, 11/6/28. https://doi.org/10.1109/SSP.2011.5967835
    Matsuyama Y, Hayashi R, Yokote R. Fast estimation of Hidden Markov Models via alpha-EM algorithm. In IEEE Workshop on Statistical Signal Processing Proceedings. 2011. p. 89-92. 5967835 https://doi.org/10.1109/SSP.2011.5967835
    Matsuyama, Yasuo ; Hayashi, Ryunosuke ; Yokote, Ryota. / Fast estimation of Hidden Markov Models via alpha-EM algorithm. IEEE Workshop on Statistical Signal Processing Proceedings. 2011. pp. 89-92
    @inproceedings{db4e4045efa241c3ad8addaee97091e8,
    title = "Fast estimation of Hidden Markov Models via alpha-EM algorithm",
    abstract = "Fast estimation algorithms of Hidden Markov Models (HMMs), or alpha-HMMs, are presented. Such novel algorithms inherit speedup properties of the alpha-EM algorithm. Since the alpha-EM algorithm includes the traditional log-EM algorithm as its special case, the alpha-HMM also includes the traditional log-HMM as its special case. This generalization appears as the utilization of the past information which is the main device of the speedup. Since the memorization of the past information requires only little increase of computational load and memory, the iteration speedup directly appears as that of CPU time. Experimental results are given.",
    keywords = "alpha-EM algorithm, alpha-HMM, past information, speedup",
    author = "Yasuo Matsuyama and Ryunosuke Hayashi and Ryota Yokote",
    year = "2011",
    doi = "10.1109/SSP.2011.5967835",
    language = "English",
    isbn = "9781457705700",
    pages = "89--92",
    booktitle = "IEEE Workshop on Statistical Signal Processing Proceedings",

    }

    TY - GEN

    T1 - Fast estimation of Hidden Markov Models via alpha-EM algorithm

    AU - Matsuyama, Yasuo

    AU - Hayashi, Ryunosuke

    AU - Yokote, Ryota

    PY - 2011

    Y1 - 2011

    N2 - Fast estimation algorithms of Hidden Markov Models (HMMs), or alpha-HMMs, are presented. Such novel algorithms inherit speedup properties of the alpha-EM algorithm. Since the alpha-EM algorithm includes the traditional log-EM algorithm as its special case, the alpha-HMM also includes the traditional log-HMM as its special case. This generalization appears as the utilization of the past information which is the main device of the speedup. Since the memorization of the past information requires only little increase of computational load and memory, the iteration speedup directly appears as that of CPU time. Experimental results are given.

    AB - Fast estimation algorithms of Hidden Markov Models (HMMs), or alpha-HMMs, are presented. Such novel algorithms inherit speedup properties of the alpha-EM algorithm. Since the alpha-EM algorithm includes the traditional log-EM algorithm as its special case, the alpha-HMM also includes the traditional log-HMM as its special case. This generalization appears as the utilization of the past information which is the main device of the speedup. Since the memorization of the past information requires only little increase of computational load and memory, the iteration speedup directly appears as that of CPU time. Experimental results are given.

    KW - alpha-EM algorithm

    KW - alpha-HMM

    KW - past information

    KW - speedup

    UR - http://www.scopus.com/inward/record.url?scp=80052231399&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=80052231399&partnerID=8YFLogxK

    U2 - 10.1109/SSP.2011.5967835

    DO - 10.1109/SSP.2011.5967835

    M3 - Conference contribution

    SN - 9781457705700

    SP - 89

    EP - 92

    BT - IEEE Workshop on Statistical Signal Processing Proceedings

    ER -