Optimized neural network search of Higgs boson production with the Tevatron

Making the best choice of kinematic variables is one of the main steps in using Neural Networks (NN) in high-energy physics. Our optimizations are based on the analysis of the Feynman diagram structure (singularities and spin effects) for the signal and background processes. Applying this method leads to improved efficiency of the Higgs search compared with the earlier NN strategy and the conventional analysis