Machine Learning for Materials Hard and Soft

ESI-DCAFM-TACO-VDSP Summer School 2022

In-person event

11 - 22 July 2022

ESI - University of Vienna
Vienna, Austria

Registration deadline: 26 April 2022

Driven by the availability of large data sets and the application of artificial intelligence algorithms, materials science is currently undergoing a dramatic transformation. In particular, machine learning approaches are opening up new possibilities to model, understand, design and discover new materials. In this summer school, geared towards doctoral and advanced master students, participants will gain an understanding of various machine learning methods and learn about their application to hard and soft materials. Introductory lectures will be complemented with hands-on exercises as well as research talks about the current state of the art.

There is no registration fee but registration is compulsory. Furthermore, MSc and PhD students have the opportunity to apply for travel support.

Places for the school are limited. It is planned that the school will take place on site. However, the school may be switched online if required due to the COVID situation.

Applications are now closed.

Main Lectures

TopicLecturer
Mathematical introduction

Philipp Grohs (University of Vienna)

Image analysis

Thomas Pock (TU Graz)

Machine Learning for electronic structure

James Spencer & David Pfau (Deep Mind)

Machine Learning potentials

Philipp Marquetand (University of Vienna)

Statistical sampling

Peter Wirnsberger (Deep Mind)

Free energies and enhanced sampling

Luigi Bonati (ETH Zurich & IIT Genova)

Materials properties prediction

Taylor D. Sparks (University of Utah)

Research Talks

Case studies

On Friday, 15 July, we will welcome Machine Learning experts who translated their knowledge into the private sector.

Schedule

The full schedule can be found here.

Organization

The ESI-DCAFM-TACO-VDSP Summer School 2022 "Machine Learning for Materials Hard and Soft" is organized by Christoph Dellago (University of Vienna), Ulrike Diebold (TU Wien), Leticia González (University of Vienna) and Jani Kotakoski (University of Vienna). Financial support by the FWF (doc.funds HiDHyS, SFB TACO), ESI and the VDSP is highly appreciated.