Speakers

Image © Aalto University, used under license

Confirmed Speakers

(In alphabetical order)

The talk schedule can be accessed from the Schedule tab.

Alice Allen

Los Alamos National Laboratory, USA

"Learning Together: Towards foundational models for machine learning interatomic potentials with meta-learning"

Alice Allen is a postdoctoral researcher at Los Alamos National Laboratory. Prior to this, she was a postdoctoral research associate at the University of Luxembourg and the University of Cambridge, where she also completed her PhD. Her work has focused on developing new forms of molecular potentials, including machine learning potentials.

Nongnuch Artrith

Utrecht University, The Netherlands

"Accelerated Sampling and Machine Learning Models for Non-Crystalline Energy Materials"

Nong Artrith is a tenured Assistant Professor in the Materials Chemistry and Catalysis Group at the Debye Institute for Nanomaterials Science, Utrecht University, and a Visiting Researcher at Microsoft Research Amsterdam Lab.  Prior to joining Utrecht University, Nong was a Research Scientist at Columbia University, USA, and a PI in the Columbia Center for Computational Electrochemistry. Nong obtained her PhD in Theoretical Chemistry from Ruhr University Bochum, Germany, for the development of machine-learning (ML) models for materials chemistry.  She was awarded a Schlumberger Foundation fellowship for postdoctoral research at MIT and subsequently joined UC Berkeley as an associate specialist.  In 2019, Nong has been named a Scialog Fellow for Advanced Energy Storage.  Since 2023, Nong is a member of the NL ARC CBBC.  She is the main developer of the open-source ML package ænet for atomistic simulations.  Her research interests focus on the development and application of first principles and ML methods for the computational discovery of energy materials and for the interpretation of experimental observations.

Jörg Behler

Ruhr-Universität Bochum, Germany

"Four Generations of Machine Learning Potentials" (online presentation)

Jörg Behler graduated in chemistry at the University of Dortmund in 2000. In 2004 he obtained his PhD at the Fritz-Haber-Institute in Berlin. After a postdoctoral stay at the ETH Zürich, in 2007 he established his own research group at the Ruhr-Universität Bochum funded by a Liebig, an Emmy Noether and a Heisenberg fellowship. In 2013 he received the Hans G. A. Hellmann award for his work on the development of high-dimensional neural network potentials. In 2017 he became a full professor for theoretical chemistry at the University of Göttingen. In 2022 he returned to Bochum for a research professorship at the newly founded Research Center Chemical Sciences and Sustainability and for establishing a new Chair for Theoretical Chemistry. His main research interest is the development and application of machine learning potentials in chemistry and materials sciences.

Michele Ceriotti

École polytechnique fédérale de Lausanne, Switzerland

"Equivariant representations: (why) do we need them?" (online presentation)

Michele Ceriotti received his Ph.D. in Physics from ETH Zürich. He spent three years in Oxford as a Junior Research Fellow at Merton College. Since 2013 he leads the laboratory for Computational Science and Modeling, in the institute of Materials at EPFL, that focuses on method development for atomistic materials modelling based on statistical mechanics and machine learning. Rather than about grants and prizes, he would like to brag that he still finds time to contribute to the development of several open-source software packages, including i-PI and chemiscope, and to serve the atomistic modelling community as an associate editor of the Journal of Chemical Physics, as a moderator of the physics.chem-ph section of the arXiv, and as an editorial board member of Physical Review Materials.

Rose Cersonsky 

University of Winsconsin-Madison, USA

Title TBD

Rose joined UW-Madison Chemical and Biological Engineering as the Conway Assistant Professor in January 2023. Prior to this, she obtained her degree in Materials Science and Engineering from University of Connecticut and PhD in Macromolecular Science and Engineering at University of Michigan, and worked for three years at École Polytechnique Fédérale de Lausanne (EPFL) as a postdoctoral researcher. For her work, Rose has received several accolades, including the ACS Colloids Victor K. LaMer Dissertation Award, the Michigan Biointerfaces Innovator Award, and most recently in being named one of Matter’s 35 under 35 materials researchers.

Cecilia Clementi

Freie Universität Berlin, Germany

"Navigating protein landscapes with a machine-learned transferable coarse-grained model"

Cecilia Clementi is Einstein Professor of Physics at Freie Universität (FU) Berlin, Germany. She joined the faculty of FU in June 2020 after 19 years as a Professor of Chemistry at Rice University in Houston, Texas. Cecilia obtained her Ph.D. in Physics at SISSA and was a postdoctoral fellow at the University of California, San Diego, where she was part of the La Jolla Interfaces in Science program. Her research focuses on the development and application of methods for the modeling of complex biophysical processes, by means of molecular dynamics, statistical mechanics, coarse-grained models, experimental data, and machine learning. Cecilia's research has been recognized by a National Science Foundation CAREER Award (2004), and the Robert A. Welch Foundation Norman Hackerman Award in Chemical Research (2009). Since 2016 she is also a co-Director of the National Science Foundation Molecular Sciences Software Institute.

Edvin Fako

BASF Switzerland

“MLFF in Heterogeneous Catalysis: an Enabling Technology”

Edvin Fako is an Application Scientist with BASF AG Switzerland, working within the global discipline QM Inorganics lead by Dr. Sandip De. Edvin obtained his PhD in the group of Prof. Nuria Lopez in ICIQ Tarragona, Spain, working on single atom/site heterogeneous catalysis. Following the completed PhD, Edvin moved to BASF SE in Ludwigshafen, Germany where he started working on combining high throughput experiment and computation in heterogenous catalysis, with the focus on green and sustainable chemistry, and machine learning workflows, dataset design and curation for machine learned interatomic potentials for applications in heterogeneous catalysis.

Zheyong (Bruce) Fan

Varian Medical Systems

"Building a unified neuroevolution potential (NEP) model for 16 metals"

Dr. Zheyong Fan obtained his PhD in physics from Nanjing University in 2010, he did postdoctoral research during 2010-2019 at Xiamen University and Aalto University. He is now working at Varian Medical Systems as a research scientist. He is the major developer of the GPUMD code and has developed the efficient neuroevolution machine-learned potential.

Guillaume Fraux

École polytechnique fédérale de Lausanne, Switzerland

"A common language for atomistic machine learning models"

Guillaume Fraux did his undergrad at École Normale Supérieure in Paris, and a PhD at PSL university in Paris, studying flexible nanoporous materials and simulations methods for the deformation of materials during adsorption. He is now a postdoctoral researcher in the COSMO laboratory at EPFL in Lausanne, Switzerland, working on machine learning applications for chemistry. He works on both theoretical tools (feature space comparison, new equivariant representations, handling of chemical complexity in simulations) and software tools for machine learning. Among other software projects, he is working on a data exchange library (metatensor) providing an interface to train MLIP; fast calculation of representations (rascaline), online data visualisation (chemiscope) and classical machine learning tools (scikit-matter).

Andrea Grisafi

École Normale Supérieure, France

"Combining electron-density and long-range machine-learning methods for the study of electrochemical interfaces"

Andrea Grisafi studied Chemistry at University of Pisa and Scuola Normale Superiore of Pisa. In 2021, he received his PhD in Materials Science from École Polytechnique Fédérale de Lausanne, under the supervision of Prof. Michele Ceriotti. His thesis was focused on the development of equivariant and long-range machine-learning models for the prediction of materials properties. After that, he joined the group of Prof. Rodolphe Vuilleumier at École Normale Supérieure of Paris, where he worked on the prediction of the electronic charge transfer in metallic frameworks. Since March 2023, he joined the group of Prof. Mathieu Salanne at Sorbonne Université of Paris, where he is working on the application of long-range machine-learning methods for the study of electrochemical interfaces.

Dávid Kovács

University of Cambridge, UK

"MACE: Higher Order Equivariant Message Passing Force Fields"

Dávid Kovács is a final year PhD student in Cambridge at the group of Prof. Csanyi. His research focuses on the development of fast and accurate force fields for molecular simulations. This includes both method development and applications to challenging chemistry problems. Prior to his PhD, Dávid also researched applications of language models to chemical reaction prediction. In his free time, he likes travelling, experimenting with coffee, and analysing football games. 

Chris Pickard

University of Cambridge, UK

"Ephemeral Data Derived Potentials - Throw Away MLPs?"

Chris Pickard’s research helps the modern scientist “see” and discover the universe at the atomic scale through Quantum Mechanics – from the centres of giant exoplanets, to pharmaceutical compounds, new battery materials and high temperature superconductors. He is the inaugural Sir Alan Cottrell Professor of Materials Science in the Department of Materials Science and Metallurgy, University of Cambridge. Previously he was Professor of Physics, University College London, and Reader in Physics, University of St Andrews. He has held EPSRC Advanced and Leadership Research Fellowships, and a Royal Society Wolfson Research Merit Award. He is a lead developer of the widely used CASTEP computer code, and introduced both the GIPAW approach to the prediction of magnetic resonance parameters and Ab Initio Random Structure Searching (AIRSS). CASTEP has provided a significant source of licencing income for Cambridge Enterprise for over 20 years, while his AIRSS software is freely available through an open source license. In 2015 he won the Rayleigh Medal and Prize of the Institute of Physics, awarded for distinguished research in theoretical, mathematical or computational physics.

Martin Uhrin

Université Grenoble Alpes, France

"Equivariant machine learning: A natural and highly data-efficient tool for predicting physical quantities"

Martin Uhrin studied computational physics at Edinburgh University, where he obtained his M.Phys. (2010) and went on to complete a PhD at University College London (2015) on the topic of high-throughput structure prediction of materials. As a postdoctoral researcher at EPFL (2015-2019), Martin created the workflow engine that powers the AiiDA materials informatics platform and was responsible for machine learning research to accelerate materials discovery. After a period at DTU (2019-2021) researching metal-air batteries, Martin returned to EPFL (2021) to conduct independent research using invariant and equivalent machine learning models to enable the inverse design of materials and molecules. In November 2023, Martin will start a PI position as a research fellow chair at the Multidisciplinary Institute in Artificial Intelligence, Université Grenoble Alpes.

Sander Vandenhaute

Ghent University, Belgium

"Rare event sampling at the top of Jacob's ladder"

Sander Vandenhaute is a fourth-year PhD student in the group of Veronique Van Speybroeck at Ghent University (BE). His research interests are largely situated at the intersection of quantum chemistry and machine learning. At the moment, Sander is mainly focused on developing fast and accurate free energy calculation techniques which are generally applicable to both chemical and physical transformations (e.g. chemical reactions or phase transitions). This consists of a mix of efficient active learning workflows, enhanced sampling techniques, and high-level quantum mechanical calculations. The main deliverable so far has been psiflow; an end-to-end framework to develop interatomic potentials for rare events. Before this, Sander worked on dimensionality reduction/coarse-grained models of nanoporous materials.

Chuck Witt

University of Cambridge, UK

"Making sense of permutation- and rotation-symmetric polynomials for machine-learned interatomic potentials"

Chuck Witt is a Junior Research Fellow (Christ's College) in the Materials Theory Group at the University of Cambridge. His interests include the development of methods and software rooted in quantum-mechanical first principles, as well as their application to problems in materials science and engineering. He earned a PhD in Mechanical and Aerospace Engineering from Princeton University.

Linfeng Zhang

AI for Science Institute and DP Technology, China

"AI-assisted molecular modelling: from multi-scale to pre-trained models" (online presentation)

Linfeng Zhang, a researcher at the AI for Science Institute in Beijing and founder of DP Technology, holds a background in applied mathematics from Princeton University and physics from Peking University. His work concentrates on the interdisciplinary field of AI for Science, contributing to machine learning, computational physics and chemistry, and materials and drug design. Linfeng is the major developer of DeePMD-kit, an open-source software for molecular simulation, and has been promoting the DeepModeling community for AI for Science enthusiasts. His efforts have led to several significant projects and recognition, including the ACM Gordon Bell Prize in 2020, and a feature on the cover of Forbes Asia's 30 Under 30 list for 2022