Completed
A Unified Understanding of ML Potentials
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Unified Understanding of E(3)-Equivariant Interatomic Potentials - Theory and Applications
Automatically move to the next video in the Classroom when playback concludes
- 1 Interatomic potentials (force fields)
- 2 Symmetries in Interatomic Potentials
- 3 Machine Learning Interatomic Potentials
- 4 A Unified Understanding of ML Potentials
- 5 Message Passing Neural Networks
- 6 Body ordered messages
- 7 Generalized Atomic Cluster Expansion (ACE)
- 8 Multi-ACE: A Framework of Many-Body Equivariant MPNNS
- 9 MPNNS - ACE identification
- 10 Classifying models in the Multi-ACE framework
- 11 MPNNs as a sparsification of local models
- 12 Understanding Nequl in the unified design space
- 13 BOTNet : A body ordered Equivariant MPNN
- 14 Influence of non-linearities
- 15 Data Normalization
- 16 ML Interatomic Potentials limitations
- 17 Solving MPNNs limitations with many body messages
- 18 Required number of message passing
- 19 Higher order messages change the learning law
- 20 High accuracy on benchmarks
- 21 Data Efficiency
- 22 Extrapolation and speed
- 23 Acetyl-acetone: H transfer
- 24 Outlook
