Convolutional Deep Neural Networks on GPUs for Particle Physics Applications

Toolkit for Multivariate Analysis (TMVA) is a machine learning toolkit for the ROOT scientific software framework. It contains a variety of techniques for data analysis and prediction such as Likelihood Estimation, Support Vector Machines, Decision trees, etc. It is used mainly for particle physics applications. Currently only feed forward Deep Neural Networks with GPU support are available in the TMVA. Recently CNNs and other DNNs have proven useful in a variety of applications like classification and tracking of particles, detection of new particles, etc. The aim of this project is to expand the current library of TMVA DNN with optimized Convolutional Neural Networks (CNN)