skip to content

Materials Modelling Seminar

Professor Michele Ceriotti, Laboratory of Computational Science and Modelling, EPFL

Wednesday 24th April 2019, 12:00

Goldsmiths 1 (0_017), Department of Materials Science & Metallurgy

Title: Atomistic Machine Learning between Physics and Data


Statistical regression techniques have become very fashionable as a tool to predict the properties of systems at the atomic scale, sidestepping much of the computational cost of first-principles simulations and making it possible to perform simulations that require thorough statistical sampling without compromising on the accuracy of the electronic structure model.

In this talk I will argue how data-driven modelling can be rooted in a mathematically rigorous and physically-motivated framework, and how this is beneficial to the accuracy and the transferability of the model. I will also highlight how machine learning - despite amounting essentially to data interpolation - can provide important physical insights on the behaviour of complex systems, on the synthesizability and on the structure-property relations of materials.

I will give examples concerning all sorts of atomistic systems, from semiconductors to molecular crystals, and properties as diverse as drug-protein interactions, dielectric response of aqueous systems and NMR chemical shielding in the solid state.

Latest news

Machine learning shows how hydrogen becomes a metal inside giant planets

10 September 2020

By combining machine learning and quantum mechanics, researchers have carried out simulations to discover how hydrogen becomes a metal under extreme pressures.

MSM-AIMR Joint Online Workshop 2020

8 September 2020

In the week of 24-28 August 2020, the Department of Materials Science and Metallurgy held a Joint Online Workshop with Tohoku University in Japan.

Hierarchically Structured Allotropes of Phosphorus from Data-Driven Exploration

22 June 2020

Researchers discover new elemental phosphorus structures by using 'fragments' of phosphorus as building blocks.

New academic license for CASTEP

6 March 2020

The CASTEP developers announce a new cost-free worldwide source code license to CASTEP and NMR CASTEP for academic use.

Dr Chuck Witt joins the MTG

6 March 2020

The Materials Theory Group welcomes its newest research associate, Dr Chuck Witt.

Welcome to Dr Bartomeu Monserrat

10 January 2020

The Materials Theory Group welcomes new faculty member and new group member Bartomeu Monserrat.

CASTEP: From research code to software product with Professor Chris Pickard

9 January 2020

On 22 January 2020, Chris Pickard will give a talk followed by a Q&A session on commercialising software.