Machine learning based event classification for the energy-differential measurement of the natC(n,p) and natC(n,d) reactions

The paper explores the feasibility of using machine learning techniques, in particular neural networks, for classification of the experimental data from the joint <math display="inline" id="d1e994" altimg="si4.svg"><msup><mrow/><mrow><mtext>nat</mtext></mrow></msup></math>C(n,p) and <math display="inline" id="d1e1005" altimg="si4.svg"><msup><mrow/><mrow><mtext>nat</mtext></mrow></msup></math>C(n,d) reaction cross section measurement from the neutron time of flight facility n_TOF at CERN. Each relevant <math display="inline" id="d1e1017" altimg="si72.svg"><mrow><mi>Δ</mi><mi>E</mi><mo linebreak="goodbreak" linebreakstyle="after">−</mo><mi>E</mi></mrow></math> pair of strips from two segmented silicon telescopes is treated separately and afforded its own dedicated neural network. An important part of the procedure is a careful preparation of training datasets, based on the raw data from Geant4 simulations. Instead of using these raw data for the training of neural networks, we divide a relevant 3-parameter space into discrete voxels, classify each voxel according to a particle/reaction type and submit these voxels to a training procedure. The classification capabilities of the structurally optimized and trained neural networks are found to be superior to those of the manually selected cuts.