Machine Learning Techniques in High Energy Physics

This course covers the modern techniques of High Energy Physics(HEP)data analysis as applied to LHC experiments. Starting with a discussionof some basic statistics concepts, we then give an overview of MonteCarlo methods and will focus on Madgraph generator. After that we diveinto machine learning techniques for data analysis. Starting with simpletools like the multilayer perceptron and then moving to modern deeplearning. We show how these tools can be applied to classification andpattern recognition problems in HEP data analysis. The lectures willinclude tutorial sessions where students will have hands on experiencewith tools like the Madgraph Monte Carlo event generator and KERASwith Tensor Flow backend for Deep Learning tools.   Programme: 1-Introduction to Statistics for HEP Data Analysis:- Basic concepts of statistics- Parameter and interval estimation- Hypothesis testing and goodness of the fit2-The Physics of Event Generators:- Physics processes , Feynman diagrams and cross sections- Monte Carlo method- Madgraph event generator3-Machine Learning:- Introduction to machine learning: classification and patternrecognition problems- The multilayer perceptron (MLP)- Universal approximation , vanishing gradient and deep learning- Convolutional network, autoencoder and adversarial network