REVIEW 21 cited by
The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentials
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentials
read the original abstract
The rapid progress of machine learning interatomic potentials over the past couple of years produced a number of new architectures. Particularly notable among these are the Atomic Cluster Expansion (ACE), which unified many of the earlier ideas around atom density-based descriptors, and Neural Equivariant Interatomic Potentials (NequIP), a message passing neural network with equivariant features that showed state of the art accuracy. In this work, we construct a mathematical framework that unifies these models: ACE is generalised so that it can be recast as one layer of a multi-layer architecture. From another point of view, the linearised version of NequIP is understood as a particular sparsification of a much larger polynomial model. Our framework also provides a practical tool for systematically probing different choices in the unified design space. We demonstrate this by an ablation study of NequIP via a set of experiments looking at in- and out-of-domain accuracy and smooth extrapolation very far from the training data, and shed some light on which design choices are critical for achieving high accuracy. Finally, we present BOTNet (Body-Ordered-Tensor-Network), a much-simplified version of NequIP, which has an interpretable architecture and maintains accuracy on benchmark datasets.
Forward citations
Cited by 21 Pith papers
-
Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models
OMat24 releases a new open dataset of 110M+ DFT calculations and EquiformerV2 models achieving SOTA on Matbench Discovery with F1>0.9 for stability and 20 meV/atom accuracy for formation energies.
-
Synthesizability, hardness, and stacking order in multicomponent transition metal carbides from machine-learned potentials
Fine-tuned MACE potentials screen 1500 equiatomic multicomponent carbides, showing group number controls stability/hardness, short-range order effects are small, and a new family of stacking-ordered phases appears in ...
-
Mixture of Experts Framework in Machine Learning Interatomic Potentials for Atomistic Simulations
A co-trained multifidelity mixture-of-experts MLIP partitions simulations into high- and low-capacity regions, maintains exact energy conservation and bulk modulus alignment, and runs more than twice as fast as a sing...
-
Representations from Pretrained Machine-Learning Interatomic Potentials as Coarse Coordinates for Material Generation and Evaluation
A dual-featurizer transport distance using MACE and contrastive GNN features jointly measures quality and novelty of generated crystals, and the MACE features can condition a flow-matching generator.
-
Learning to Prepare Molecular Ground States with Transformer Models
Transformers trained on ADAPT-VQE data generate imipramine ground-state circuits in seconds at roughly reference accuracy — and beat the training data after reinforcement learning — though real-hardware energies still...
-
Anisotropic representations for E(3)-equivariant machine learning coarse-grained potentials
An E(3)-equivariant ML potential for anisotropic ellipsoidal coarse-grained beads reproduces water’s radial, angular and orientational structure far better than an isotropic baseline while delivering large speedups.
-
CoTAR: Topology and Atomic State Reconstruction in Condensed Phases
CoTAR reconstructs molecular topology from atomic positions in non-reactive condensed-phase systems using a GNN-HMM hybrid, reporting 0.906 F1 on classical MD data and raising uMLIP valid-snapshot rate from 38.6% to 8...
-
A Combined Tight Binding with Machine Learning Potential Model for Magnesium Compounds
Hybrid DFTB+MACE model for Mg systems replaces pair potential with MACE trained on DFT-DFTB differences to improve accuracy over standard DFTB while retaining electronic structure access.
-
Charting the thermodynamic stability of hybrid perovskite alloys with machine learning
A two-level ML approach using graph neural network potentials and direct energy predictors maps thermodynamic stability across (Cs/FA)Pb(Br/I)3 and (Cs/FA)Sn(Br/I)3 compositions, finding narrower stable regions for Sn...
-
Compact SO(3) Equivariant Atomistic Foundation Models via Structural Pruning
Structural pruning of SO(3) equivariant atomistic models from large checkpoints yields 1.5-4x fewer parameters and 2.5-4x less pre-training compute than small models trained from scratch, while outperforming them on m...
-
Displacive quantum critical point in superconducting hydrides: The case of H$_3$S
Path-integral simulations with a machine-learned potential place the displacive quantum critical point of H3S at ~134 GPa and associate the superconducting Tc peak with the surrounding quantum fluctuations.
-
Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models
OMat24 provides over 110 million DFT calculations and EquiformerV2 models that reach state-of-the-art performance on material stability and formation energy prediction.
-
Quantum machine learning interatomic potential: Application of variational quantum algorithm
Replacing the final layer of a pretrained ANI interatomic potential with a small quantum circuit training only the circuit parameters gives slightly lower energy RMSE than a classical layer, only where the pretrained ...
-
A Combined Tight Binding with Machine Learning Potential Model for Magnesium Compounds
A DFTB+MACE model that replaces the pairwise repulsive term with a trained many-body potential improves forces, phonons, and surface energies for MgO and related systems while retaining electronic-structure output.
-
Benchmarking machine-learned interatomic potentials for molecular infrared spectroscopy
Equivariant MPNNs (MACE, PaiNN, SO3Net) outperform invariant ones (SchNet, FieldSchNet) in transferability and spectral accuracy for IR spectroscopy of organic molecules while maintaining high fidelity on training data.
-
Direct Simulation of LiNi0.8Mn0.1Co0.1O2 Transport Properties Using an Efficient and Accurate Machine Learning Potential
A fine-tuned MACE-based machine learning potential trained via active learning on near-ground-state DFT data enables direct large-scale MD prediction of lithium self-diffusion coefficients in NMC811.
-
Systematic Fine-Tuning of MACE Interatomic Potentials for Catalysis
Fine-tuned MACE MLIPs achieve lower mean absolute errors on catalytic reaction energies and barriers than from-scratch models, with a large fine-tuned model performing best on both metallic and oxide systems including...
-
Electron dynamics mediate the water-carbon {\pi} bond
Electron dynamics in the aromatic pi cloud quench and amplify specific vibrational signals from a single water molecule bound to pyrene anion.
-
Machine-Learned Force Fields for Lattice Dynamics at Coupled-Cluster Level Accuracy
Machine-learned force fields trained on coupled-cluster potential energy surfaces produce phonon dispersions and vibrational densities of states for solids that agree better with experiment than DFT-based models.
-
Machine Learning Interatomic Potentials: Advancing Open-Source Software for Efficient and Scalable Molecular Simulation
mlip v2 is a new software release that integrates API redesign, e3j backend, eSEN model, improved charge modeling, and expanded simulation capabilities to support larger-scale molecular modeling.
-
Comparing fine-tuning strategies of MACE machine learning force field for modeling Li-ion diffusion in LiF for batteries
MACE-MPA-0 predicts Li diffusion Ea of 0.22 eV in LiF, fine-tuned version with 300 points gives 0.20 eV, close to DeePMD reference of 0.24 eV, using far less training data.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.