MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields.
Ilyes BatatiaDávid Péter KovácsGregor N. C. SimmChristoph OrtnerGábor CsányiPublished in: NeurIPS (2022)
Keyphrases
- message passing
- higher order
- neural network
- markov random field
- belief propagation
- distributed systems
- shared memory
- factor graphs
- approximate inference
- probabilistic inference
- graphical models
- pairwise
- sum product
- inference in graphical models
- artificial neural networks
- high quality
- conditional random fields
- message passing interface
- clique potentials
- distributed shared memory
- probability distribution
- graph cuts
- ldpc codes
- max product
- probabilistic model
- multiscale
- computer vision