MoMa: A Modular Deep Learning Framework for Material Property Prediction
This addresses the challenge of handling diverse material tasks for researchers in materials science, representing a novel paradigm rather than an incremental improvement.
The paper tackles the problem of material property prediction by introducing MoMa, a modular deep learning framework that adaptively composes specialized modules for diverse tasks, achieving a 14% average improvement over baselines across 17 datasets.
Deep learning methods for material property prediction have been widely explored to advance materials discovery. However, the prevailing pre-train then fine-tune paradigm often fails to address the inherent diversity and disparity of material tasks. To overcome these challenges, we introduce MoMa, a Modular framework for Materials that first trains specialized modules across a wide range of tasks and then adaptively composes synergistic modules tailored to each downstream scenario. Evaluation across 17 datasets demonstrates the superiority of MoMa, with a substantial 14% average improvement over the strongest baseline. Few-shot and continual learning experiments further highlight MoMa's potential for real-world applications. Pioneering a new paradigm of modular material learning, MoMa will be open-sourced to foster broader community collaboration.