Method home for mlp-mc-1

In this method a fully connected multilayer perceptron neural network with a single hidden layer of hyperbolic tangent units is trained using Bayesian learning implemented via Markov Chain Monte Carlo (MCMC) techniques. For a more detailed theoretical account of the method, refer to Radford Neal: Bayesian Learning for Neural Networks, Springer Verlag, New York (1996).


The software used to implement this method is a very general package of Markov Chain sampling techniques for Bayesian learning in neural networks developed by Radford Neal.

The definition of the mlp-mc-1 method is given in a postscript document which also includes a sample script (written for the csh shell) to be used in conjunction with the software.


Directory listing of the results available for the mlp-mc-1 method. Put the desired files in the appropriate methods directory in your delve hierarchy and uncompress them with using the "gunzip *.gz" command and untar them using "tar -xvf *.tar".

Related References

A similar method was considered in: Carl Edward Rasmussen "A Practical Monte Carlo Implementation of Bayesian Learning", Advances in Neural Information Processing Systems 8, eds. D. S. Touretzky, M. C. Mozer, M. E. Hasselmo, MIT Press, 1996.

Last Updated by Carl Edward Rasmussen, September 16, 1996