Machine Learning Introduction

The day will provide delegates with a chance to discuss the landscape of machine learning including the range of algorithms on the market, the interplay between developments in academia and industry, and the future of machine learning. Throughout the day, students will learn about supervised learning for classification problems, specifically using random forests, decision trees, and neural networks.

Students will develop machine learning classifiers in Python to analyse proton-proton collision data from the ATLAS experiment at the Large Hadron Collider to search for new physics. These examples will be a springboard from which students can explore machine learning in other applications. Dedicated discussion will also consider the benefits and drawbacks of machine learning techniques, including their ethical issues and practical limitations, and how these can mitigated. By the end of the day, delegates should have acquired an elementary understanding of how to use and optimise machine learning in a scientifically robust way.