AI Applications through the Whole Life Cycle of Material Discovery

Abstract

We provide a review of machine learning (ML) tools for material discovery and sophisticated applications of different ML strategies. Although there have been a few published reviews on artificial intelligence (AI) for materials with an emphasis on a single material system or individual methods, this paper focuses on an applicationbased perspective in AI-enhanced material discovery. It shows how AI strategies are applied through material discovery stages (including characterization, property prediction, synthesis, and theory paradigm discovery). Also, by referring to the ML tutorial, readers can acquire a better understanding of the exact functions of ML methods in each application and how these methods work to realize the targets. We are aiming to enable a better integration of AI methods with the material discovery process. The keys to successful applications of AI in material discovery and challenges to be addressed are also highlighted.

Toward autonomous design and synthesis of novel inorganic materials

“…As will be discussed below, active learning techniques are gaining traction throughout materials science, with multiple applications recently demonstrated in automated workflows [101][102][103][104] .…”

Characterising soft matter using machine learning

“…Using machine learning in the design of complex materials at the atomic level has been explored extensively in recent years. [100][101][102] This includes research to optimise specific properties of crystalline materials via iterating between experiments in the lab and the generation of refined computational suggestions. 103 In this context, a cost function is being minimised for which each new ''function evaluation'' involves fabricating a new sample.…”

Self‐Driven Multistep Quantum Dot Synthesis Enabled by Autonomous Robotic Experimentation in Flow

“…Generally, experimentation strategies with high reliance on user guidance and operation, require increased time commitment and level of expertise. Recent advances in AI, including deep neural networks (DNN) and reinforcement learning, for rapid chemical space exploration, [49][50][51][52][53][54][55][56][57][58][59][60][61] provide an exciting opportunity to reshape the development and on-demand manufacturing of colloidal nanomaterials. Consequently, a number of self-optimizing microfluidic reactors have been developed to take advantage of continuous material exploration in a low chemical consumption system.…”

“…The self-optimizing fluidic studies frequently apply either established optimization methods, such as Stable Noisy Optimization by Branch and Fit, [62] or recently re-emerged Gaussian process regression-based modeling (e.g., Kriging), [36,63] are often placed in tandem with techniques that use decision making under uncertainty in a closed-loop fashion, such as the case of many recent efforts in the use of Bayesian Optimization (BO) for autonomous materials development. [56,64] In recent work, we demonstrated the use of ensemble neural networks (ENNs) for application in a fully self-optimizing fluidic system. [31] While the most effective AI algorithm is problem-dependent, ENN-based experimentation methods have shown their potential in navigating large reaction spaces from a position of no prior knowledge as well as with archived data as a starting position to achieve a specific objective.…”