Spiers Memorial Lecture: How to do impactful research in artificial intelligence for chemistry and materials science

Abstract

Machine learning has been pervasively touching many fields of science. Chemistry and materials science are no exception. While machine learning has been making a great impact, it is still not reaching its full potential or maturity. In this perspective, we first outline current applications across a diversity of problems in chemistry. Then, we discuss how machine learning researchers view and approach problems in the field. Finally, we provide our considerations for maximizing impact when researching machine learning for chemistry.

References in corpus (72)

• XGBoost: A Scalable Tree Boosting System
• Automatic chemical design using a data-driven continuous representation of molecules
• Hierarchical Text-Conditional Image Generation with CLIP Latents
• DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
• Learning to Plan Chemical Syntheses
• Scaling Laws for Neural Language Models
• Molecular Transformer - A Model for Uncertainty-Calibrated Chemical Reaction Prediction
• Molecular Contrastive Learning of Representations via Graph Neural Networks
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• GuacaMol: Benchmarking Models for De Novo Molecular Design
• The Open Catalyst 2020 (OC20) Dataset and Community Challenges
• Retrieval-Augmented Generation for Large Language Models: A Survey
• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
• ChatGPT Chemistry Assistant for Text Mining and Prediction of MOF Synthesis
• Deep neural network solution of the electronic Schrödinger equation
• Ab-Initio Solution of the Many-Electron Schrödinger Equation with Deep Neural Networks
• ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction
• Neural networks for the prediction organic chemistry reactions
• Machine learning for electronically excited states of molecules
• Autonomous discovery in the chemical sciences part I: Progress
• The Open Catalyst 2022 (OC22) Dataset and Challenges for Oxide Electrocatalysts
• Autonomous discovery in the chemical sciences part II: Outlook
• Measuring Mathematical Problem Solving With the MATH Dataset
• Chemical Shifts in Molecular Solids by Machine Learning
• Galactica: A Large Language Model for Science
• A foundation model for atomistic materials chemistry
• Coarse Graining Molecular Dynamics with Graph Neural Networks
• Gryffin: An algorithm for Bayesian optimization of categorical variables informed by expert knowledge
• Machine learning in nuclear materials research
• AdsorbML: A Leap in Efficiency for Adsorption Energy Calculations using Generalizable Machine Learning Potentials
• The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery
• MatterSim: A Deep Learning Atomistic Model Across Elements, Temperatures and Pressures
• Excited state, non-adiabatic dynamics of large photoswitchable molecules using a chemically transferable machine learning potential
• Group SELFIES: A Robust Fragment-Based Molecular String Representation
• MatterGen: a generative model for inorganic materials design
• Predicting charge density distribution of materials using a local-environment-based graph convolutional network
• Flow-matching -- efficient coarse-graining of molecular dynamics without forces
• The Impact of Large Language Models on Scientific Discovery: a Preliminary Study using GPT-4
• Accurate Computation of Quantum Excited States with Neural Networks
• Learning Pair Potentials using Differentiable Simulations
• A Hitchhiker's Guide to Geometric GNNs for 3D Atomic Systems
• Matbench Discovery -- A framework to evaluate machine learning crystal stability predictions
• Bayesian Optimization of Catalysis With In-Context Learning
• Language models can generate molecules, materials, and protein binding sites directly in three dimensions as XYZ, CIF, and PDB files
• ORGANA: A Robotic Assistant for Automated Chemistry Experimentation and Characterization
• GFlowNets for AI-Driven Scientific Discovery
• Are large language models superhuman chemists?
• Errors are Useful Prompts: Instruction Guided Task Programming with Verifier-Assisted Iterative Prompting
• LAB-Bench: Measuring Capabilities of Language Models for Biology Research
• Inverse molecular design and parameter optimization with Hückel theory using automatic differentiation
• Less can be more for predicting properties with large language models
• A Review of Large Language Models and Autonomous Agents in Chemistry
• Genetic algorithms are strong baselines for molecule generation
• GP-MoLFormer: A Foundation Model For Molecular Generation
• RGFN: Synthesizable Molecular Generation Using GFlowNets
• Reflection-Equivariant Diffusion for 3D Structure Determination from Isotopologue Rotational Spectra in Natural Abundance
• FraGNNet: A Deep Probabilistic Model for Tandem Mass Spectrum Prediction
• From Text to Insight: Large Language Models for Materials Science Data Extraction
• Transferable Boltzmann Generators
• Navigating protein landscapes with a machine-learned transferable coarse-grained model
• Efficient Evolutionary Search Over Chemical Space with Large Language Models
• End-to-End Crystal Structure Prediction from Powder X-Ray Diffraction
• The ab initio amorphous materials database: Empowering machine learning to decode diffusivity
• RePLan: Robotic Replanning with Perception and Language Models
• A Recipe for Charge Density Prediction
• React-OT: Optimal Transport for Generating Transition State in Chemical Reactions
• ChemGymRL: An Interactive Framework for Reinforcement Learning for Digital Chemistry
• Thermodynamically Informed Multimodal Learning of High-Dimensional Free Energy Models in Molecular Coarse Graining
• Imagen 3
• Accurate and efficient structure elucidation from routine one-dimensional NMR spectra using multitask machine learning
• Physics-Informed Neural Networks and Beyond: Enforcing Physical Constraints in Quantum Dissipative Dynamics

Cited by in corpus (1)

• All-in-one foundational models learning across quantum chemical levels