Talks.cam will close on 1 July 2026, further information is available on the UIS Help Site
 

University of Cambridge > Talks.cam > SciSoc – Cambridge University Scientific Society > Machine learning force fields shows extreme generalisation

Machine learning force fields shows extreme generalisation

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact ajaf3.

I will introduce the general problem of first principles force fields: creating surrogate models for quantum mechanics that yield the energy of a configuration of atoms in 3D space, as we would find them in materials or molecules. Over the last decade significant advances were made in the attainable accuracy, and today we can model materials and molecules with a per-atom energy accuracy of up to 1 part in 10,000 with a speedup of over a million or more compared to the explicit quantum mechanical calculation, enabling accurate molecular dynamics simulations on large length and time scales. The most surprising aspect of the best models is its extreme generalisation: fitted only on small periodic crystals, it shows stable trajectories on arbitrary chemical systems, from water to nanoparticles and proteins. The precise relationship between the architectural elements and the extreme generalisation is still a mystery. The locality of the graph neural network structure is key to its success, as well as high body order and message passing. The force fields get significantly better with more data, yet model size and complexity can remain largely the same. Current challenges include integrating explicit long range electrostatics and combining large datasets for materials and organic molecules where the appropriate levels of electronic structure theory are incompatible.

Talks are priced at £4 for non-Scientific Society members. Scientific Society members will have free access to all our talks. Lifetime membership costs £15 and gives free access to all talks, members-only events and priority access to oversubscribed SciSoc events.

This talk is part of the SciSoc – Cambridge University Scientific Society series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity