The Open Catalyst 2020 (OC20) Dataset and Community Challenges

Abstract

Catalyst discovery and optimization is key to solving many societal and energy challenges including solar fuels synthesis, long-term energy storage, and renewable fertilizer production. Despite considerable effort by the catalysis community to apply machine learning models to the computational catalyst discovery process, it remains an open challenge to build models that can generalize across both elemental compositions of surfaces and adsorbate identity/configurations, perhaps because datasets have been smaller in catalysis than related fields. To address this we developed the OC20 dataset, consisting of 1,281,040 Density Functional Theory (DFT) relaxations (~264,890,000 single point evaluations) across a wide swath of materials, surfaces, and adsorbates (nitrogen, carbon, and oxygen chemistries). We supplemented this dataset with randomly perturbed structures, short timescale molecular dynamics, and electronic structure analyses. The dataset comprises three central tasks indicative of day-to-day catalyst modeling and comes with pre-defined train/validation/test splits to facilitate direct comparisons with future model development efforts. We applied three state-of-the-art graph neural network models (CGCNN, SchNet, Dimenet++) to each of these tasks as baseline demonstrations for the community to build on. In almost every task, no upper limit on model size was identified, suggesting that even larger models are likely to improve on initial results. The dataset and baseline models are both provided as open resources, as well as a public leader board to encourage community contributions to solve these important tasks.

arXiv comment37 pages, 11 figures, submitted to ACS Catalysis

References in corpus (15)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Fast Graph Representation Learning with PyTorch Geometric
• VQA: Visual Question Answering
• Directional Message Passing for Molecular Graphs
• Beyond scaling relations for the description of catalytic materials
• Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules
• Machine Learning in Thermodynamics: Prediction of Activity Coefficients by Matrix Completion
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Predicting charge density distribution of materials using a local-environment-based graph convolutional network
• Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties
• Prediction of Atomization Energy Using Graph Kernel and Active Learning
• A community-powered search of machine learning strategy space to find NMR property prediction models
• A High-Throughput Solver for Marginalized Graph Kernels on GPU

Cited by in corpus (64)

• Recent Advances and Applications of Deep Learning Methods in Materials Science
• DeePMD-kit v2: A software package for Deep Potential models
• Towards Universal Neural Network Potential for Material Discovery Applicable to Arbitrary Combination of 45 Elements
• The Open Catalyst 2022 (OC22) Dataset and Challenges for Oxide Electrocatalysts
• WILDS: A Benchmark of in-the-Wild Distribution Shifts
• DPA-2: a large atomic model as a multi-task learner
• OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphs
• AdsorbML: A Leap in Efficiency for Adsorption Energy Calculations using Generalizable Machine Learning Potentials
• A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
• On the redundancy in large material datasets: efficient and robust learning with less data
• Crystal Diffusion Variational Autoencoder for Periodic Material Generation
• AI-driven inverse design of materials: Past, present and future
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Probing out-of-distribution generalization in machine learning for materials
• Graph Neural Networks Accelerated Molecular Dynamics
• JARVIS-Leaderboard: A Large Scale Benchmark of Materials Design Methods
• Predicting binding motifs of complex adsorbates using machine learning with a physics-inspired graph representation
• Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic potentials
• Rotation Invariant Graph Neural Networks using Spin Convolutions
• Generative Language Model for Catalyst Discovery
• Recent progress in the JARVIS infrastructure for next-generation data-driven materials design
• BOTAN: BOnd TArgeting Network for prediction of slow glassy dynamics by machine learning relative motion
• On the design space between molecular mechanics and machine learning force fields
• A Quantum-Chemical Bonding Database for Solid-State Materials
• Spherical Message Passing for 3D Graph Networks
• Large-Scale Representation Learning on Graphs via Bootstrapping
• Automated 3D Pre-Training for Molecular Property Prediction
• An end-to-end attention-based approach for learning on graphs
• Machine Learning and Data-Driven Methods in Computational Surface and Interface Science
• Discovery of stable surfaces with extreme work functions by high-throughput density functional theory and machine learning
• Navigating chemical reaction space with a steering wheel
• ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations
• Ligand additivity relationships enable efficient exploration of transition metal chemical space
• Geometry-enhanced Pre-training on Interatomic Potentials
• Accelerating the prediction of inorganic surfaces with machine learning interatomic potentials
• Efficient first principles based modeling via machine learning: from simple representations to high entropy materials
• Machine Learning Accelerated Descriptor Design for Catalyst Discovery in CO${}_2$ to Methanol Conversion
• An Introduction to Electrocatalyst Design using Machine Learning for Renewable Energy Storage
• Spiers Memorial Lecture: How to do impactful research in artificial intelligence for chemistry and materials science
• Generator of Neural Network Potential for Molecular Dynamics: Constructing Robust and Accurate Potentials with Active Learning for Nanosecond-scale Simulations
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Boosting Heterogeneous Catalyst Discovery by Structurally Constrained Deep Learning Models
• AI and Quantum Computing in Binary Photocatalytic Hydrogen Production
• Augmentation of Universal Potentials for Broad Applications
• Geometric and Physical Quantities Improve E(3) Equivariant Message Passing
• Facilitating {\it ab initio} configurational sampling of multicomponent solids using an on-lattice neural network model and active learning
• Multitask methods for predicting molecular properties from heterogeneous data
• Scalable Training of Trustworthy and Energy-Efficient Predictive Graph Foundation Models for Atomistic Materials Modeling: A Case Study with HydraGNN
• Beyond Independent Error Assumptions in Large GNN Atomistic Models
• The JARVIS Infrastructure is All You Need for Materials Design
• Understanding Discrepancies of Wavefunction Theories for Large Molecules
• Towards Foundation Models for Materials Science: The Open MatSci ML Toolkit
• Crystalline Material Discovery in the Era of Artificial Intelligence
• Rational design of nanoscale stabilized oxide catalysts for OER with OC22
• $\Delta$-model correction of Foundation Model based on the models own understanding
• Coarse-grained crystal graph neural networks for reticular materials design
• AI for Scientific Discovery is a Social Problem
• mkite: A distributed computing platform for high-throughput materials simulations
• Ab initio machine learning in chemical compound space
• Transition States Energies from Machine Learning: An Application to Reverse Water-Gas Shift on Single-Atom Alloys
• Comparing the latent features of universal machine-learning interatomic potentials
• CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning
• Bridging Text and Crystal Structures: Literature-driven Contrastive Learning for Materials Science
• Atoms to Processes: The Role of Artificial Intelligence and Machine Learning in Chemical Engineering