Using deep neural network to reduce mismatch in ALICE Time of Fligth

The Time Of Flight (TOF) belong to the ALICE experiment and is dedicated to particle identification by computing their time of fligh. It works by matching a hit and a track reconstructed by the inner detectors. During that procedure there is some hit/track mismatch that can be reduced by using machine learning techniques. The first part of the internship was dedicated to the study of the problem with an old ROOT library called TMultiLayerPerceptron in order to improve PID with TOF. The second part was dedicated to the use of Keras, a much more recent python library.