Learning with First , Second , and No Derivatives : aCase Study in High Energy

In this paper diierent algorithms for training multi-layer perceptron architecture are applied to a signiicant discrimination task in High Energy Physics. The One-Step Secant technique is compared with On-Line Backpropagation, the \Bold Driver" batch version and Conjugate Gradient methods. In addition, a new algorithm (AAne Shaker) is proposed that uses stochastic search based on function values and aane transformations of the local search region. Although the AAne Shaker requires more CPU time to reach the maximum generalization, the technique can be interesting for special-purpose VLSI implementations and for non-diierentiable functions.