Development of LSTM and GRU layers in TMVA

This project is about development of Long Short Term Memory(LSTM) and Gated Recurrent Unit(GRU) layers in TMVA, both of which belong to a general class of neural networks called the Recurrent Neural Networks(RNN). These layers have many important applications in the realm of data analysis for particle physics experiments. As an example, LSTMs can be used for track reconstruction of charged particles in the Large Hadron Collider(LHC). They can also be used for analyzing the voltage time series from the electronic monitoring system present in superconducting LHC magnets.