New Algorithm to Investigate Neural Networks

Random cost simulations were introduced as a method to investigate optimization problems in systems with con icting constraints. Here I study the approach in connection with the training of a feed-forward multilayer perceptron, as used in high energy physics applications. It is suggested to use random cost simulations for generating a set of selected con gurations. On each of those nal minimization may then be performed by a standard algorithm. For the training example at hand many almost degenerate local minima are thus found. Some e ort is spend to discuss whether they lead to equivalent classi cations of the data. This research was partially funded by the Department of Energy under contract DE-FG05-87ER40319. Department of Physics, The Florida State University, Tallahassee, FL 32306, USA. Supercomputer Computations Research Institute, Tallahassee, FL 32306, USA. E-mail: [email protected]