High Energy Physics Seminars

Machine Learning in Particle Physics (1/3)

by Prof. Harrison B Prosper (Florida State University, USA)

Thursday, March 15, 2018 from to (Asia/Kolkata)
at TIFR, Mumbai ( D-406 )
Machine learning has been used in particle physics since the later half of 1980s and most recently in the discovery of the Higgs boson in 2012. However, in spite of its widespread use in particle physics, machine learn- ing has yet to shed its reputation of being inscrutable. To be sure, there is a bewildering array of methods. But, when viewed with sufficient detachment it becomes clear that what one is really doing with these methods is fitting highly flexible, high-dimensional, functions to data in order to estimate functions, often probabilities. In these lectures, the focus is on the ideas and methods in machine learning that are most relevant to particle physics. But, I also briefly touch upon a few state- of-the-art methods that have revolutionized applications such as face recognition and autonomous navigation, and which may yet find routine application in particle physics.