Accelerating Materials Discovery: Learning a Universal Representation of Chemical Processes for Cross-Domain Property Prediction
This addresses the bottleneck of heterogeneous data in materials discovery for researchers, though it is incremental as it builds on existing graph neural network methods.
The paper tackles the problem of slow and costly experimental validation in materials discovery by introducing a universal directed-tree process-graph representation that unifies heterogeneous data, and their model, trained on 700,000 process graphs from 9,000 documents, achieves strong performance when fine-tuned on domain-specific datasets with minimal additional data.
Experimental validation of chemical processes is slow and costly, limiting exploration in materials discovery. Machine learning can prioritize promising candidates, but existing data in patents and literature is heterogeneous and difficult to use. We introduce a universal directed-tree process-graph representation that unifies unstructured text, molecular structures, and numeric measurements into a single machine-readable format. To learn from this structured data, we developed a multi-modal graph neural network with a property-conditioned attention mechanism. Trained on approximately 700,000 process graphs from nearly 9,000 diverse documents, our model learns semantically rich embeddings that generalize across domains. When fine-tuned on compact, domain-specific datasets, the pretrained model achieves strong performance, demonstrating that universal process representations learned at scale transfer effectively to specialized prediction tasks with minimal additional data.