Learning Local Equivariant Representations for Large-Scale Atomistic Dynamics | Albert Musaelian
Try datamol.io - the open source toolkit that simplifies molecular processing and featurization workflows for machine learning scientists working in drug discovery: https://datamol.io/
Never miss another M2D2 talk, add the schedule to your calendar: https://m2d2.io/talks/m2d2/about/
Also consider joining the M2D2 Slack: https://m2d2group.slack.com/join/shar...
Abstract: Trade-offs between accuracy and speed have long limited the applications of machine learning interatomic potentials. Recently, E(3)-equivariant architectures have demonstrated leading accuracy, data efficiency, transferability, and simulation stability, but their computational cost and scaling has generally reinforced this trade-off. In particular, the ubiquitous use of message passing architectures has precluded the extension of accessible length- and time-scales with efficient multi-GPU calculations.
In this talk I will discuss Allegro, a strictly local equivariant deep learning interatomic potential designed for parallel scalability and increased computational efficiency that simultaneously exhibits excellent accuracy. After presenting the architecture, I will discuss applications and benchmarks on various materials and chemical systems, including recent demonstrations of scaling to large all-atom biomolecular systems such as solvated proteins and a 44 million atom model of the HIV capsid. Finally, I will summarize the software ecosystem and tooling around Allegro.
Speaker:
Albert Musaelian - https://www.krellinst.org/csgf/fellow...
Twitter Prudencio: / tossouprudencio
Twitter Jonny: / hsu_jonny
Twitter datamol.io: / datamol_io
~
Chapters:
00:00 - Intro
06:01 - Machine Learning Potentials
11:24 - No Constraints, Invariance, Equivariance
15:55 - NequIP Generalizes Across Geometry
20:04 - Message Passing Networks
21:09 - Allegro: Strictly Local Deep Equivariant Model
35:16 - Importance of Locality
39:54 - Demonstrating Allegro Scaling Up
49:56 - Weak Scaling
54:56 - Q+A