Journal article icon

Journal article

Sequential Monte Carlo Methods to Train Neural Network Models

Abstract:

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 sequential Monte Carlo techniques. The new algorithm may be viewed as a global optimization strategy that allows us to learn the probability distributions of the network weights and outputs in a sequential...

Expand abstract

Actions


Access Document


Publisher copy:
10.1162/089976600300015664

Authors


De Freitas More by this author
M.A. Niranjan More by this author
Journal:
Neural Computation
Volume:
12
Issue:
4
Pages:
955-993
Publication date:
2000
DOI:
ISSN:
0899-7667
URN:
uuid:335f89d9-df1f-4849-94c9-f94f1d5db20a
Local pid:
cs:7544

Terms of use


Metrics



If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP