MODULAR NEURAL-NET WORKS FOR ONLINE EVENT CLASSIFICATION IN HIGH-ENERGY PHYSICS

We discuss the application of Modular Neural Networks (MNNs) for high- performance, high rate classification of HEP events, both in terms of the algorithms involved, and their hardware implementation. Three diff erent problems were treated successfully with the MNN framework, namel y the identification of electrons and photons in the first and second trigger levels of the CMS experiment and the classification of Cherenk ov rings in a RICH detector, showing the versatility and conceptual si mplicity of MNN for triggering in HEP experiments. A prototype of the electron/photon trigger primitives generation system for the CMS exper iment, based on a MNN and implemented with L-Neuro 2.3 chips: was deve loped and tested in the CERN SPS H4 beam line, The system reached exce llent performance, identifying electrons with full efficiency at the 4 0 MHz LHC clock. The same hardware, properly reprogrammed, is able to handle the trigger of strangelet rings in the context of an experiment to search for exotic matter. (C) 1998 Elsevier Science B.V. All right s reserved.