Alert button
Picture for Stefano Curtarolo

Stefano Curtarolo

Alert button

Physics in the Machine: Integrating Physical Knowledge in Autonomous Phase-Mapping

Nov 15, 2021
A. Gilad Kusne, Austin McDannald, Brian DeCost, Corey Oses, Cormac Toher, Stefano Curtarolo, Apurva Mehta, Ichiro Takeuchi

Figure 1 for Physics in the Machine: Integrating Physical Knowledge in Autonomous Phase-Mapping
Figure 2 for Physics in the Machine: Integrating Physical Knowledge in Autonomous Phase-Mapping

Application of artificial intelligence (AI), and more specifically machine learning, to the physical sciences has expanded significantly over the past decades. In particular, science-informed AI or scientific AI has grown from a focus on data analysis to now controlling experiment design, simulation, execution and analysis in closed-loop autonomous systems. The CAMEO (closed-loop autonomous materials exploration and optimization) algorithm employs scientific AI to address two tasks: learning a material system's composition-structure relationship and identifying materials compositions with optimal functional properties. By integrating these, accelerated materials screening across compositional phase diagrams was demonstrated, resulting in the discovery of a best-in-class phase change memory material. Key to this success is the ability to guide subsequent measurements to maximize knowledge of the composition-structure relationship, or phase map. In this work we investigate the benefits of incorporating varying levels of prior physical knowledge into CAMEO's autonomous phase-mapping. This includes the use of ab-initio phase boundary data from the AFLOW repositories, which has been shown to optimize CAMEO's search when used as a prior.

Viaarxiv icon

On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning

Jun 11, 2020
A. Gilad Kusne, Heshan Yu, Changming Wu, Huairuo Zhang, Jason Hattrick-Simpers, Brian DeCost, Suchismita Sarker, Corey Oses, Cormac Toher, Stefano Curtarolo, Albert V. Davydov, Ritesh Agarwal, Leonid A. Bendersky, Mo Li, Apurva Mehta, Ichiro Takeuchi

Figure 1 for On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
Figure 2 for On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
Figure 3 for On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
Figure 4 for On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning

Active learning - the field of machine learning (ML) dedicated to optimal experiment design, has played a part in science as far back as the 18th century when Laplace used it to guide his discovery of celestial mechanics [1]. In this work we focus a closed-loop, active learning-driven autonomous system on another major challenge, the discovery of advanced materials against the exceedingly complex synthesis-processes-structure-property landscape. We demonstrate autonomous research methodology (i.e. autonomous hypothesis definition and evaluation) that can place complex, advanced materials in reach, allowing scientists to fail smarter, learn faster, and spend less resources in their studies, while simultaneously improving trust in scientific results and machine learning tools. Additionally, this robot science enables science-over-the-network, reducing the economic impact of scientists being physically separated from their labs. We used the real-time closed-loop, autonomous system for materials exploration and optimization (CAMEO) at the synchrotron beamline to accelerate the fundamentally interconnected tasks of rapid phase mapping and property optimization, with each cycle taking seconds to minutes, resulting in the discovery of a novel epitaxial nanocomposite phase-change memory material.

* 30 pages and 13 figures in PDF including Methods section 
Viaarxiv icon

Machine learning modeling of superconducting critical temperature

Oct 06, 2017
Valentin Stanev, Corey Oses, A. Gilad Kusne, Efrain Rodriguez, Johnpierre Paglione, Stefano Curtarolo, Ichiro Takeuchi

Figure 1 for Machine learning modeling of superconducting critical temperature
Figure 2 for Machine learning modeling of superconducting critical temperature
Figure 3 for Machine learning modeling of superconducting critical temperature
Figure 4 for Machine learning modeling of superconducting critical temperature

Superconductivity has been the focus of enormous research effort since its discovery more than a century ago. Yet, some features of this unique phenomenon remain poorly understood; prime among these is the connection between superconductivity and chemical/structural properties of materials. To bridge the gap, several machine learning schemes are developed herein to model the critical temperatures ($T_{\mathrm{c}}$) of the 12,000+ known superconductors available via the SuperCon database. Materials are first divided into two classes based on their $T_{\mathrm{c}}$ values, above and below 10 K, and a classification model predicting this label is trained. The model uses coarse-grained features based only on the chemical compositions. It shows strong predictive power, with out-of-sample accuracy of about 92%. Separate regression models are developed to predict the values of $T_{\mathrm{c}}$ for cuprate, iron-based, and "low-$T_{\mathrm{c}}$" compounds. These models also demonstrate good performance, with learned predictors offering potential insights into the mechanisms behind superconductivity in different families of materials. To improve the accuracy and interpretability of these models, new features are incorporated using materials data from the AFLOW Online Repositories. Finally, the classification and regression models are combined into a single integrated pipeline and employed to search the entire Inorganic Crystallographic Structure Database (ICSD) for potential new superconductors. We identify more than 30 non-cuprate and non-iron-based oxides as candidate materials.

* npj Computational Materials 4, Article number: 29 (2018)  
* 17 pages, 7 figures 
Viaarxiv icon