You are currently reviewing an older revision of this page.
2017 AFOSR MURIScientific Autonomous Reasoning Agent (SARA): Integrating Materials Theory, Experiment and ComputationPO: Dr. Richard Riecken, Science of Information, Computation, Learning, and FusionLead-PI: Robert Bruce van Dover, Cornell UniversityMURI Website
Our vision is to integrate materials science and computer science methods to dramatically accelerate, by orders of magnitude, the discovery and development of new materials. We will demonstrate how cross-disciplinary collaboration involving materials scientists and computer scientists can significantly advance and enrich the state of the art of each discipline.
Materials discovery and development provides unique challenges and opportunities for the next generation of Artificial Intelligent (AI) systems. We have recently witnessed an exponential growth in the generation of theoretical and experimental materials data. The extraction of actionable knowledge from this rich data has emerged as a grand challenge that invites not only application but also advancement of the state-of-the-art in computational science. Yet there is a substantial barrier to the integration of computation and materials science due to the need for domain-specific knowledge and incorporation of heterogeneous data. These challenges offer opportunities to make foundational advancements in both fields; materials development incorporating theory and experiment is the ideal platform for addressing these grand challenges.
We propose to develop the Scientific Autonomous Reasoning Agent (SARA), a multi-Agent system that accelerates materials discovery and development—from identifying promising new candidate materials to guiding their synthesis and establishing their value for solving prominent materials challenges. SARA will integrate, in a synergistic and complementary way, first principles quantum physics, experimental materials synthesis, processing, and characterization, and AI based algorithms for reasoning and conducting science, including the representation, planning, optimization, and learning of materials knowledge. The hierarchical multi-agent system generates hypotheses and autonomously tests them through autonomous execution of the materials discovery/development process, which is enabled through concomitant development of robotic processing and characterization tools, on-the-fly density functional theory (DFT) calculations, and AI-based algorithms. SARA is further augmented with human insights so that the artificial intelligence leverages that of expert scientists, creating an unprecedented platform for human-machine collaboration that embodies a marked advancement in DoD capabilities. SARA’s autonomous materials discovery/development platform includes methodologies for the rational design of functional materials and for discovering the requisite synthesis parameters for both stable and metastable materials. SARA will bring to fruition the vision of the Materials Genome Initiative, enhance DoD proficiency in solving materials challenges, and usher in a new era of scientific discovery.