This module will give an overview of Machine Learning (ML) and its methodologies and examples of applications. As an hors d'oeuvre, we will make a transition from statistics to machine learning using regression models. Then we will discover the beauty and power of deep neural networks - one of the most flexible approaches to supervised learning. Unsupervised Learning will free us from labeled data, as an application we look at clustering. The last method we will discover is reinforcement learning. A powerful method in which a machine or an agent interacts with its environment, performs actions, and learns by a trial-and-error method. Imagine a computer playing chess against itself many times, using trial and error strategy to learn. For all of the methods Python codes will be made available, in order to support curiosity driven exploration of this fascinating field.
Short bio Andreas Adelmann
Andreas Adelmann is a senior scientist and head of the interdisciplinary Laboratory for Scientific Computing and Modeling with close to 30 scientists at the Paul Scherrer Institut in Switzerland. At ETH, he teaches courses in accelerator modelling, computational physics and leads a seminar on computational physics. Dr. Adelmann obtained a PhD in applied mathematics from the ETH Zurich on the subject of numerical modelling of high-power cyclotrons. His research interests include advanced accelerator concepts, non-linear dynamics, large-scale optimisation, high- performance-computing and machine learning. In 2001, Dr. Adelmann was the first recipient of the Alvarez fellowship in computational science awarded by the Lawrence Berkeley Laboratory. In 2012, together with researchers from ETH Zurich and IBM Research Zurich, he received the PRACE award, recognising a breakthrough in science achieved with high-performance computing resources in the area of reduced-order modelling and optimisation.
Massimo Giovannozzi / Participants: 130 (Zoom) - 35 (in presence)