The Bayesian analysis of neural networks is difficult because the prior over functions has a complex form, leading to implementations that either make approximations or use Monte Carlo integration techniques. In this paper I investigate the use of Gaussian process priors over functions, which permit the predictive Bayesian analysis to be carried out exactly using matrix operations. The method has been tested on two challenging problems and has produced excellent results.
|Title of host publication||Mathematics of Neural Networks|
|Subtitle of host publication||Models, Algorithms and Applications|
|Number of pages||5|
|Publication status||Published - 1995|
|Name||Operations Research/Computer Science Interfaces Series|