Using software and hardware neural networks in a Higgs search

The present investigation uses information from computer simulations to train neural networks to identify decays of heavy Higgs particles (mH ⪢ mz). Results are presented both for software and hardware analog neural networks. The hardware tests include the Intel ETANN and the CLNN32/CLNS64 (experimental, research prototype developed at Bellcore) chip-set implemented in VME-modules. The processing and learning times for the networks are discussed.