Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials
Machine learning interatomic potentials (MLIPs) have become a hallmark of AI for scientific simulation. While efforts on new architectures and datasets have led to increasingly accurate and general models, the choice of optimizer for training has largely remained unexplored, defaulting to Adam and its variants in the community. Here, we implement and systematically compare a class of recently proposed matrix-structured optimizers, including Muon, SOAP, and the hybrid SOAP-Muon, for training NequIP and Allegro MLIP models. We find that these optimizers can substantially outperform Adam in both
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 54%iree-org/iree →
- PossiblePossibly related (embedding) · 54%Introducing GeneBench-Pro →
- PossiblePossibly related (embedding) · 53%DeepSeek open-sources inference optimizations with 60–85% faster generation [pdf] →
- PossiblePossibly related (embedding) · 52%JuliaDecisionFocusedLearning/InferOpt.jl →
- PossiblePossibly related (embedding) · 51%tensorflow/serving →
- LinkedLinked via arxiv author · 85%Gil Harari →
“Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials”
- LinkedLinked via arxiv author · 85%Yoel Zimmermann →
“Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials”
- LinkedLinked via arxiv author · 85%Ola Tangen Kulseng →
“Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials”