Fast α-weighted EM learning for neural networks of module mixtures

Yasuo Matsuyama, Satoshi Furukawa, Naoki Takeda, Takayuki Ikeda

    研究成果: Conference contribution

    6 被引用数 (Scopus)

    抄録

    A class of extended logarithms is used to derive α-weighted EM (α-weighted Expectation and Maximization) algorithms. These extended EM algorithms (WEM's, α-EM's) have been anticipated to outperform the traditional (logarithmic) EM algorithm on the speed. The traditional approach falls into a special case of the new WEM. In this paper, general theoretical discussions are given first. Then, clear-cut evidences that show faster convergence than the ordinary EM approach are given on the case of mixture-of-expert neural networks. This process takes three steps. The first step is to show concrete algorithms. Then, the convergence is theoretically checked. Thirdly, experiments on the mixture-of-expert learning are tried to show the superiority of the WEM. Besides the supervised learning, unsupervised case on a Gaussian mixture is also examined. Faster convergence of the WEM is observed again.

    本文言語English
    ホスト出版物のタイトルIEEE International Conference on Neural Networks - Conference Proceedings
    編集者 Anon
    Place of PublicationPiscataway, NJ, United States
    出版社IEEE
    ページ2306-2311
    ページ数6
    3
    出版ステータスPublished - 1998
    イベントProceedings of the 1998 IEEE International Joint Conference on Neural Networks. Part 1 (of 3) - Anchorage, AK, USA
    継続期間: 1998 5 41998 5 9

    Other

    OtherProceedings of the 1998 IEEE International Joint Conference on Neural Networks. Part 1 (of 3)
    CityAnchorage, AK, USA
    Period98/5/498/5/9

    ASJC Scopus subject areas

    • Software

    フィンガープリント 「Fast α-weighted EM learning for neural networks of module mixtures」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

    引用スタイル