AI4Science Events

Biweekly Lab Meetings

The AI4Science Lab Meetings are a series of bi-weekly research and literature presentations, held on alternating Thursdays at 11:00 hr, starting this academic year on 11 September. All members take a turn to present their work or a thought-provoking paper. The meetings take place at Science Park and are not online. If you wish to take part in our Lab Meetings, send an email to us with your name, affiliation and a one-sentence motivation for joining.


AI4Science Colloquium

Coming AI4Science colloquia.

Title: Foundational models for materials chemistry

Speaker: Professor Gábor Csány, Department of Engineering, University of Cambridge, United Kingdom.

  • Date: Monday 1 December 2025
  • Time: 13:00 hr
  • Location: Science Park, room tba

Abstract:

A new computational task has been defined and solved over the past 15 years for extended material systems: the analytic fitting of the Born-Oppenheimer potential energy surface as a function of nuclear coordinates. The resulting potentials (“force fields”) are reactive, many-body, with evaluation costs that are currently on the order of 0.1-10 ms/atom/cpu core (or about 1-10ms on a powerful GPU), and reach accuracies of a few meV/atom when trained specifically for a given system using iterative or active learning methods. The latest and most successful architectures leverage many-body symmetric descriptions of local geometry and equivariant message passing networks. Perhaps the most surprising recent result is the stability of models trained on very diverse training sets across the whole periodic table. Our recent discovery is that the MACE-MP-0 model that was trained on just ~150,000 real and hypothetical small inorganic crystals (90% of training set < 70 atoms), is capable of stable molecular dynamics at ambient conditions on any system tested so far - this includes crystals, liquids, surfaces, clusters, molecules, and combinations of all of these. The astounding generalisation performance of such foundation models open the possibility to creating a universally applicable interatomic potential with useful accuracy for materials (especially when fine-tuned with a little bit of domain-specific data), and democratise quantum-accurate large scale molecular simulations by lowering the barrier to entry into the field. Similarly, in the domain of organic chemistry, training just on small molecules and small clusters allows accurate simulation of condensed phase systems, and first principles prediction of quantities such as hydration free energies for the first time.

Biography

Gábor Csányi is a Hungarian-born scientist at the University of Cambridge. He read mathematics in Cambridge as an undergraduate at St. John’s College and did his PhD work in physics at MIT. After postdoctoral research in the Cavendish Laboratory, he joined the faculty in the Department of Engineering at Cambridge. He won an F. W. Bessel award from the Alexander von Humboldt foundation.

His work is in the field of computational chemistry, and is focused on developing algorithms to predict the properties of materials and molecules from first principles. He pioneered the application of machine learning to molecular modeling which lead to enormous gains in the efficiency of molecular dynamics simulation.

Previous Colloquia


News and events archive

Check out recent news and events on our archive. Click on the items below for further reading.

11 July 2025 › ChemAI registration now open!
4 July 2025 › AI4Science get together
Internal meeting to get to know our new AI4Science members and friends
Read More ›


23 May 2025 › AI4Physics lecture series
Journal club style tutorials on deep learning architecturs for scientific applications in physics
Read More ›


12 June 2025 › PhD defence of David Ruhe
Structured Deep Learning with Applications in Astrophysics
Read More ›


5 June 2025 › PhD defence of Fiona Lippert
From weather radars to bird migration fluxes; Process-guided machine learning for spatio-temporal forecasting and inference
Read More ›


29 November 2024 › ChemAI Symposium
Entering the fifth paradigm for chemistry
Read More ›


25 - 29 November 2024 › ChemAI Week
Webinars, research lectures, tutorials, lab tours; a week long events on the topic of Chemistry and AI!
Read More ›


15 November 2024 › PhD defence of Benjamin Kurt Miller
Machine learning for scientific simulation: Inference and generative models
Read More ›


8 November 2024 › PhD defence of Jim Boelrijk
Bayesian optimization for liquid chromatography separations
Read More ›


18 October 2024 › Data Science Day
AI for Science: opportunities and challenges
Read More ›


30 May 2024 › Kickoff Meeting
Artificial Intelligence for Sustainable Materials and Molecules
Read More ›


23 November 2023 › ChemAI Day
Entering the fifth paradigm for chemistry
Read More ›


15 May 2023 › AI4 Enzyme Instructed Catalysis Workshop
Designing synthetic catalyst systems
Read More ›


24 February 2023 › AI4Science Symposium
1 December 2022 › Symposium Chem Together #4
Chemistry meets Artificial Intelligence
Read More ›


8 July 2020 › AI4Science Kickoff Workshop