Editorial
Perspective on AI for accelerated materials design at the AI4Mat-2023 workshop at NeurIPS 2023
The AI4Mat-2023 organizing committee showcases the major developments as well as ongoing research challenges where innovative solutions can bring transformative changes to the state-of-the-art in applying AI for accelerated materials design.
Communication
A message passing neural network for predicting dipole moment dependent core electron excitation spectra
A message-passing neural network using a unit direction vector in addition to molecular graphs as the input satisfying invariance to space-inversion symmetry operations enables prediction of the anisotropic core electron excitation spectra.
Communication
Discovery of novel reticular materials for carbon dioxide capture using GFlowNets
GFlowNets discover reticular materials with simulated CO2 working capacity outperforming all materials in CoRE2019.
Paper
CoDBench: a critical evaluation of data-driven models for continuous dynamical systems
We introduce CoDBench, an exhaustive benchmarking suite comprising 12 state-of-the-art data-driven models for solving differential equations, including 4 distinct categories of models, against 10 widely applicable benchmark datasets encompassing challenges from fluid and solid mechanics.
Paper
Towards equilibrium molecular conformation generation with GFlowNets
GFlowNets allow for sampling diverse, thermodynamically feasible molecular conformations from the Boltzmann distribution.
Paper
Reconstructing the materials tetrahedron: challenges in materials information extraction
Quantifying challenges towards information extraction from scientific articles to complete the materials science tetrahedron.
Paper
Gotta be SAFE: a new framework for molecular design
SAFE is a novel SMILES-compatible, fragment-based molecular line notation that streamlines molecule generation tasks. Unlike existing line notations, it enforces a sequential depiction of molecular substructures, thus simplifying molecule design.
Paper
EGraFFBench: evaluation of equivariant graph neural network force fields for atomistic simulations
EGraFFBench: a framework for evaluating equivariant graph neural network force fields on dynamic atomistic simulations.
Paper
Learning conditional policies for crystal design using offline reinforcement learning
Conservative Q-learning for band-gap conditioned crystal design with DFT evaluations – the model is trained on trajectories constructed from crystals in the Materials Project. Results indicate promising performance for lower band gap targets.
Paper
Connectivity optimized nested line graph networks for crystal structures
Graph neural networks (GNNs) have been applied to a large variety of applications in materials science and chemistry. We report a nested line-graph neural network achieving state-of-the-art performance in multiple benchmarks.
About this collection
The AI for Accelerated Materials Design (AI4Mat) workshop at NeurIPS 2023 featured many of the ongoing major research themes in materials design, synthesis, and characterization by bringing together an international interdisciplinary community of researchers and enthusiasts. The AI4Mat 2023 organizing committee and the editors of Digital Discovery have curated a selection of research papers drawn from some of the most exciting and high-quality paper submissions from the workshop. We are pleased to share these papers, and a perspective on the workshop as a whole, in this themed collection.