Convergent on-line algorithms for supervised learning in neural networks | IEEE Journals & Magazine | IEEE Xplore

Convergent on-line algorithms for supervised learning in neural networks


Abstract:

We define online algorithms for neural network training, based on the construction of multiple copies of the network, which are trained by employing different data blocks...Show More

Abstract:

We define online algorithms for neural network training, based on the construction of multiple copies of the network, which are trained by employing different data blocks. It is shown that suitable training algorithms can be defined, in a way that the disagreement between the different copies of the network is asymptotically reduced, and convergence toward stationary points of the global error function can be guaranteed. Relevant features of the proposed approach are that the learning rate must be not necessarily forced to zero and that real-time learning is permitted.
Published in: IEEE Transactions on Neural Networks ( Volume: 11, Issue: 6, November 2000)
Page(s): 1284 - 1299
Date of Publication: 30 November 2000

ISSN Information:

PubMed ID: 18249854

References

References is not available for this document.