OMatG flash speeds up discovery of inorganic materials with fewer steps
OMatG-flash: An All-Atom Flow Map with Reinforce Adjoint Matching for Scalable Materials Discovery
Machine Learning
Summary
Finding new inorganic materials is important for technology like computers and batteries, but it's often slow and expensive to predict and generate these materials digitally. The authors created OMatG-flash, a new AI method that predicts and generates crystal structures faster and with fewer computing steps than previous methods. It uses a special technique called Reinforce Adjoint Matching to improve accuracy after training. This makes it easier and quicker to discover candidate materials for further testing and development.
What this means in practice
- •For materials engineering teams: Generate high-quality candidate inorganic materials faster to accelerate experimental testing and development.
- •For energy storage developers: Speed up the design of novel battery materials by quickly proposing promising inorganic structures.
Authors
Thomas Egg, Harry Winston Sullivan, Ellad B. Tadmor, Stefano Martiniani
Abstract
The discovery of novel inorganic materials drives technological breakthroughs in critical fields such as computing and energy storage. Generative AI has promised to accelerate the materials discovery pipeline, but state-of-the-art flow and diffusion models remain bottlenecked by the cost of proposing candidate materials. To address this, we introduce OMatG-flash, an all-atom flow map for inorganic crystal structure prediction (CSP) and de novo generation (DNG). OMatG-flash is a Pareto-optimal inference engine for materials, sampling candidate materials with an order of magnitude fewer inference steps and less wall-clock time than existing flow and diffusion models while demonstrating benchmark performance on par with the state-of-the-art. To enable post-training fine-tuning we apply Reinforce Adjoint Matching to flow maps, further improving match rates and RMSE on the unconditional CSP task. OMatG-flash showcases the potential of flow maps to accelerate generation of high-quality candidate inorganic materials and demonstrates a step forward in sample throughput necessary for data-hungry materials discovery workflows.