Sequential Monte Carlo Methods to Train Neural Network Models
We discuss a novel strategy for training neural networks using sequential Monte Carlo algorithms and propose a new hybrid gradient descent / sampling importance resampling algorithm (HySIR). In terms of computational time and accuracy, the hybrid SIR is a clear improvement over conventional sequenti...
主要な著者: | Freitas, D, Nando, Niranjan, M, Gee, A, Doucet, A |
---|---|
フォーマット: | Journal article |
出版事項: |
2000
|
類似資料
-
Sequential monte carlo methods To train neural network models
著者:: , d, 等
出版事項: (2000) -
Sequential Monte Carlo methods for diffusion processes
著者:: Jasra, A, 等
出版事項: (2009) -
Sequential Monte Carlo samplers
著者:: Del Moral, P, 等
出版事項: (2006) -
Sequential Monte Carlo for model selection and estimation of neural networks
著者:: Andrieu, C, 等
出版事項: (2000) -
Controlled sequential Monte Carlo
著者:: Heng, J, 等
出版事項: (2020)