Steering an Active Learning Workflow Towards Novel Materials Discovery via Queue Prioritization
This work addresses the inefficiency of generative AI in inverse design for materials science, specifically for carbon capture, by integrating active learning to improve candidate quality, though it is incremental as it builds on existing workflows.
The authors tackled the problem of generative AI workflows wasting resources on low-quality candidates in materials discovery by proposing a queue prioritization algorithm that combines generative modeling with active learning. Their approach increased the average number of high-performing novel molecular candidates for carbon capture from 281 to 604 out of 1000.
Generative AI poses both opportunities and risks for solving inverse design problems in the sciences. Generative tools provide the ability to expand and refine a search space autonomously, but do so at the cost of exploring low-quality regions until sufficiently fine tuned. Here, we propose a queue prioritization algorithm that combines generative modeling and active learning in the context of a distributed workflow for exploring complex design spaces. We find that incorporating an active learning model to prioritize top design candidates can prevent a generative AI workflow from expending resources on nonsensical candidates and halt potential generative model decay. For an existing generative AI workflow for discovering novel molecular structure candidates for carbon capture, our active learning approach significantly increases the number of high-quality candidates identified by the generative model. We find that, out of 1000 novel candidates, our workflow without active learning can generate an average of 281 high-performing candidates, while our proposed prioritization with active learning can generate an average 604 high-performing candidates.