Orb: A Fast, Scalable Neural Network Potential

2024/10/29 by M. Neumann, Mark Neumann, James Gin +15 · 1 voice · 51 citations
Computer Science · Physics and Astronomy · #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning (cs.LG) #Materials Science (cond-mat.mtrl-sci) #Neural Networks and Applications #cond-mat.mtrl-sci #cs.LG

paper · pdf · doi:10.48550/arxiv.2410.22570

openalex publication_date 2024/10/29 · arxiv published 2024/10/29 · arxiv updated 2024/10/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We introduce Orb, a family of universal interatomic potentials for atomistic modelling of materials. Orb models are 3-6 times faster than existing universal potentials, stable under simulation for a range of out of distribution materials and, upon release, represented a 31% reduction in error over other methods on the Matbench Discovery benchmark. We explore several aspects of foundation model development for materials, with a focus on diffusion pretraining. We evaluate Orb as a model for geometry optimization, Monte Carlo and molecular dynamics simulations.

Cited by

• Faster Molecular Dynamics with Neural Network Potentials via Distilled Multiple Time-Stepping and Nonconservative Forces
• The Convergence Frontier: Integrating Machine Learning and High Performance Quantum Computing for Next-Generation Drug Discovery
• AI-Driven Expansion and Application of the Alexandria Database
• Predicting interstitial elements in Refractory Complex Concentrated Alloys
• Evaluation of Foundational Machine Learned Interatomic Potentials for Migration Barrier Predictions
• Mofasa: A Step Change in Metal-Organic Framework Generation
• Evaluating Mechanical Property Prediction across Material Classes using Molecular Dynamics Simulations with Universal Machine-Learned Interatomic Potentials
• Li-P-S Electrolyte Materials as a Benchmark for Machine-Learned Interatomic Potentials
• Fine-Tuning Unifies Foundational Machine-learned Interatomic Potential Architectures at ab initio Accuracy
• Atomistic Simulations of Oxide-Water Interfaces using Machine Learning Potentials
• Efficient small-cell sampling for machine-learning potentials of multi-principal element alloys
• Optimizing Cross-Domain Transfer for Universal Machine Learning Interatomic Potentials
• Fine-Tuning Bulk-oriented Universal Interatomic Potentials for Surfaces: Accuracy, Efficiency, and Forgetting Control
• ADAPT: Lightweight, Long-Range Machine Learning Force Fields Without Graphs
• Accelerating Crystal Structure Prediction Using Data-Derived Potentials: High-Pressure Binary Hydrides
• Scalable Foundation Interatomic Potentials via Message-Passing Pruning and Graph Partitioning
• MLIP Arena: Advancing Fairness and Transparency in Machine Learning Interatomic Potentials via an Open, Accessible Benchmark Platform
• OMNI-P2x universal neural network potential for excited-state simulations
• Universal Machine Learning Potentials under Pressure
• A universal machine learning model for the electronic density of states
• Benchmarking Universal Interatomic Potentials on Zeolite Structures
• Training a Foundation Model for Materials on a Budget
• LeMat-Traj: A Scalable and Unified Dataset of Materials Trajectories for Atomistic Modeling
• Universal Machine Learning Potential for Systems with Reduced Dimensionality
• BLIPs: Bayesian Learned Interatomic Potentials
• Autonomous thermodynamically informed database generation for machine-learned interatomic potentials and application to magnesium
• Energy Underprediction from Symmetry in Machine-Learning Interatomic Potentials
• Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence
• Benchmark of Approximate Quantum Chemical and Machine Learning Potentials for Biochemical Proton Transfer Reactions
• MOFSimBench: Evaluating Universal Machine Learning Interatomic Potentials In Metal--Organic Framework Molecular Modeling
• Implicit Machine Learning Force Fields Accelerate Molecular Dynamics Simulations
• UMA: A Family of Universal Models for Atoms
• Leveraging neural network interatomic potentials for a foundation model of chemistry
• Machine Learning Potentials for Alloys: A Detailed Workflow to Predict Phase Diagrams and Benchmark Accuracy
• Machine learning potentials for modeling alloys across compositions
• Universally Accurate or Specifically Inadequate? Stress‐Testing General Purpose Machine Learning Interatomic Potentials
• A Graph Neural Network for the Era of Large Atomistic Models
• Benchmarking Universal Machine Learning Interatomic Potentials for Real-Time Analysis of Inelastic Neutron Scattering Data
• Cross-disciplinary perspectives on the potential for artificial intelligence across chemistry
• Flexible MOF Generation with Torsion-Aware Flow Matching
• HORM: A Large Scale Molecular Hessian Database for Optimizing Reactive Machine Learning Interatomic Potentials
• FlashMD: long-stride, universal prediction of molecular dynamics
• Accelerating point defect photo-emission calculations with machine learning interatomic potentials
• Prediction and Experimental Verification of Electrolyte Solvation Structure from an OMol25-Trained Interatomic Potential
• Pushing the limits of unconstrained machine-learned interatomic potentials
• Egret-1: Pretrained Neural Network Potentials for Efficient and Accurate Bioorganic Simulation
• MLIP-MC: A Framework for Adsorption Simulations using Machine-Learned Interatomic Potentials
• Simulations of High Temperature Decomposition of Metal-Organic Frameworks to form Amorphous Catalysts
• Molecular Learning Dynamics
• MatterTune: An Integrated, User-Friendly Platform for Fine-Tuning Atomistic Foundation Models to Accelerate Materials Simulation and Discovery
• Orb-v3: atomistic simulation at scale
• Electronic Structure Guided Inverse Design Using Generative Models
• Dispersion-corrected Machine Learning Potentials for 2D van der Waals Materials
• Cross-functional transferability in universal machine learning interatomic potentials

Discussions

• More info on Orb the code is here: github.com/orbital-mate... and the preprint is here: arxiv.org/abs/2410.22570 [bsky, 3 points, 0 comments]

Related

• Orb-v3: atomistic simulation at scale
• Universal Machine Learning Potential for Systems with Reduced Dimensionality
• Constructing high‐dimensional neural network potentials: A tutorial review
• Beyond potentials: integrated machine-learning models for materials
• NEP89: Universal neuroevolution potential for inorganic and organic materials across 89 elements