Emulating Expert Insight: A Robust Strategy for Optimal Experimental Design

The challenge of optimal design of experiments (DOE) pervades materials science, physics, chemistry, and biology. Bayesian optimization has been used to address this challenge in vast sample spaces, although it requires framing experimental campaigns through the lens of maximizing some observable. This framing is insufficient for epistemic research goals that seek to comprehensively analyze a sample space, without an explicit scalar objective (e.g., the characterization of a wafer or sample library). In this work, we propose a flexible formulation of scientific value that recasts a dataset of input conditions and higher-dimensional observable data into a continuous, scalar metric. Intuitively, the scientific value function measures where observables change significantly, emulating the perspective of experts driving an experiment, and can be used in collaborative analysis tools or as an objective for optimization techniques. We demonstrate this technique by exploring simulated phase boundaries from different observables, autonomously driving a variable temperature measurement of a ferroelectric material, and providing feedback from a nanoparticle synthesis campaign. The method is seamlessly compatible with existing optimization tools, can be extended to multi-modal and multi-fidelity experiments, and can integrate existing models of an experimental system. Because of its flexibility, it can be deployed in a range of experimental settings for autonomous or accelerated experiments.

[1]  Brian L. DeCost,et al.  Self-driving Multimodal Studies at User Facilities , 2023, Acta Crystallographica Section A Foundations and Advances.

[2]  Phillip M. Maffettone,et al.  Delivering real-time multi-modal materials analysis with enterprise beamlines , 2022, Cell Reports Physical Science.

[3]  Anthony Degennaro,et al.  Machine Learning for analysis of speckle dynamics: quantification and outlier detection , 2022, Physical Review Research.

[4]  Phillip M. Maffettone,et al.  Advancing Discovery with Artificial Intelligence and Machine Learning at NSLS-II , 2022, Synchrotron Radiation News.

[5]  A. Kusne,et al.  Scalable multi-agent lab framework for lab optimization , 2022, Matter.

[6]  Brian L. DeCost,et al.  Reproducible sorbent materials foundry for carbon capture at scale , 2022, Cell Reports Physical Science.

[7]  Paul Bendich,et al.  Topological Simplification of Signals for Inference and Approximate Reconstruction , 2022, 2023 IEEE Aerospace Conference.

[8]  Bruce Ravel,et al.  Machine learning enabling high-throughput and remote operations at large-scale user facilities , 2022, Digital Discovery.

[9]  A. Henson,et al.  Discovering New Chemistry with an Autonomous Robotic Platform Driven by a Reactivity-Seeking Neural Network , 2021, ACS central science.

[10]  I. Takeuchi,et al.  On-the-fly autonomous control of neutron diffraction via physics-informed Bayesian active learning , 2021, Applied Physics Reviews.

[11]  Aaron Stein,et al.  Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities , 2021, Nature Reviews Physics.

[12]  Semion K. Saikin,et al.  Autonomous experimentation systems for materials development: A community perspective , 2021 .

[13]  Aidan C. Daly,et al.  Constrained non-negative matrix factorization enabling real-time insights of in situ and high-throughput experiments , 2021, Applied Physics Reviews.

[14]  Phillip M. Maffettone,et al.  Gaming the beamlines—employing reinforcement learning to maximize scientific outcomes at large-scale user facilities , 2021, Mach. Learn. Sci. Technol..

[15]  A. Butté,et al.  Machine Learning for Biologics: Opportunities for Protein Engineering, Developability, and Formulation , 2021, Trends in Pharmacological Sciences.

[16]  P. Midgley,et al.  Revisiting metal fluorides as lithium-ion battery cathodes , 2021, Nature Materials.

[17]  Brian L. DeCost,et al.  On-the-fly closed-loop materials discovery via Bayesian active learning , 2020, Nature Communications.

[18]  Daniel Olds,et al.  Outlook for artificial intelligence and machine learning at the NSLS-II , 2020, Mach. Learn. Sci. Technol..

[19]  D. Olds Synchrotron X-ray Diffraction for Energy and Environmental Materials: The Current Role and Future Directions of Total Scattering Beamlines in the Functional Material Scientific Ecosystem , 2020, Synchrotron Radiation News.

[20]  Phillip M. Maffettone,et al.  Crystallography companion agent for high-throughput materials discovery , 2020, Nature Computational Science.

[21]  Jiagen Li,et al.  Autonomous discovery of optically active chiral inorganic perovskite nanocrystals through an intelligent cloud lab , 2020, Nature Communications.

[22]  David T. Jones,et al.  Improved protein structure prediction using potentials from deep learning , 2020, Nature.

[23]  Brian L. DeCost,et al.  A high-throughput structural and electrochemical study of metallic glass formation in Ni-Ti-Al. , 2019, ACS combinatorial science.

[24]  Sorelle A. Friedler,et al.  Experiment Specification, Capture and Laboratory Automation Technology (ESCALATE): a software pipeline for automated chemical experimentation and data management , 2019, MRS Communications.

[25]  Alán Aspuru-Guzik,et al.  Next-Generation Experimentation with Self-Driving Laboratories , 2019, Trends in Chemistry.

[26]  Peter I. Frazier,et al.  A Tutorial on Bayesian Optimization , 2018, ArXiv.

[27]  P. F. Peterson,et al.  Combinatorial appraisal of transition states for in situ pair distribution function analysis , 2017 .

[28]  Ryan P. Adams,et al.  Multi-Task Bayesian Optimization , 2013, NIPS.

[29]  S. Ounpraseuth,et al.  Gaussian Processes for Machine Learning , 2008 .

[30]  Appendix to: BOTORCH: A Framework for Efficient Monte-Carlo Bayesian Optimization , 2021 .

[31]  Nando de Freitas,et al.  Taking the Human Out of the Loop: A Review of Bayesian Optimization , 2016, Proceedings of the IEEE.

[32]  Sidney Addelman,et al.  trans-Dimethanolbis(1,1,1-trifluoro-5,5-dimethylhexane-2,4-dionato)zinc(II) , 2008, Acta crystallographica. Section E, Structure reports online.