このアイテムのアクセス数: 13

このアイテムのファイル:
ファイル 記述 サイズフォーマット 
1742-6596/699/1/012019.pdf1.45 MBAdobe PDF見る/開く
タイトル: Stochastic gradient method with accelerated stochastic dynamics
その他のタイトル: International Meeting on High-Dimensional Data-Driven Science (HD3-2015)
著者: Ohzeki, Masayuki
著者名の別形: 大関, 真之
発行日: Mar-2016
出版者: Institute of Physics Publishing
誌名: Journal of Physics: Conference Series
巻: 699
論文番号: 012019
抄録: We implement the simple method to accelerate the convergence speed to the steady state and enhance the mixing rate to the stochastic gradient Langevin method. The ordinary stochastic gradient method is based on mini-batch learning for reducing the computational cost when the amount of data is extraordinary large. The stochasticity of the gradient can be mitigated by the injection of Gaussian noise, which yields the stochastic Langevin gradient method; this method can be used for Bayesian posterior sampling. However, the performance of the stochastic Langevin gradient method depends on the mixing rate of the stochastic dynamics. In this study, we propose violating the detailed balance condition to enhance the mixing rate. Recent studies have revealed that violating the detailed balance condition accelerates the convergence to a stationary state and reduces the correlation time between the samplings. We implement this violation of the detailed balance condition in the stochastic gradient Langevin method and test our method for a simple model to demonstrate its performance.
著作権等: © Published under licence by IOP Publishing Ltd. Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
URI: http://hdl.handle.net/2433/218815
DOI(出版社版): 10.1088/1742-6596/699/1/012019
出現コレクション:学術雑誌掲載論文等

アイテムの詳細レコードを表示する

Export to RefWorks


出力フォーマット 


このリポジトリに保管されているアイテムはすべて著作権により保護されています。