TRI Authors: Patrick K Herring, Muratahan Aykol

All Authors: Peter M Attia, Aditya Grover, Norman Jin, Kristen A Severson, Todor M Markov, Yang-Hung Liao, Michael H Chen, Bryan Cheong, Nicholas Perkins, Zi Yang, Patrick K Herring, Muratahan Aykol, Stephen J Harris, Richard D Braatz, Stefano Ermon, William C Chueh

Simultaneously optimizing many design parameters in time-consuming experiments causes bottlenecks in a broad range of scientific and engineering disciplines1,2. One such example is process and control optimization for lithium-ion batteries during materials selection, cell manufacturing and operation. A typical objective is to maximize battery lifetime; however, conducting even a single experiment to evaluate lifetime can take months to years3,4,5. Furthermore, both large parameter spaces and high sampling variability3,6,7 necessitate a large number of experiments. Hence, the key challenge is to reduce both the number and the duration of the experiments required. Here we develop and demonstrate a machine learning methodology  to efficiently optimize a parameter space specifying the current and voltage profiles of six-step, ten-minute fast-charging protocols for maximizing battery cycle life, which can alleviate range anxiety for electric-vehicle users8,9. We combine two key elements to reduce the optimization cost: an early-prediction model5, which reduces the time per experiment by predicting the final cycle life using data from the first few cycles, and a Bayesian optimization algorithm10,11, which reduces the number of experiments by balancing exploration and exploitation to efficiently probe the parameter space of charging protocols. Using this methodology, we rapidly identify high-cycle-life charging protocols among 224 candidates in 16 days (compared with over 500 days using exhaustive search without early prediction), and subsequently validate the accuracy and efficiency of our optimization approach. Our closed-loop methodology automatically incorporates feedback from past experiments to inform future decisions and can be generalized to other applications in battery design and, more broadly, other scientific domains that involve time-intensive experiments and multi-dimensional design spaces. Read More

Citation: Attia, Peter M., Aditya Grover, Norman Jin, Kristen A. Severson, Todor M. Markov, Yang-Hung Liao, Michael H. Chen et al. "Closed-loop optimization of fast-charging protocols for batteries with machine learning." Nature 578, no. 7795 (2020): 397-402.

TRI Authors: Joseph Montoya, Jens Hummelshoj, Muratahan Aykol All Authors: Montoya, Joseph H., Kirsten Winther, Raul A. Flores, Thomas Bligaard, Jens Strabo Hummelshøj, and Muratahan Aykol We present an end-to-end computational system for autonomous materials discovery. The system aims for cost-effective optimization in large, high-dimensional search spaces of materials by adopting a sequential, agent-based approach to deciding which experiments to carry out. In choosing next experiments, agents can make use of past knowledge, surrogate models, logic, thermodynamic or other physical constructs, heuristic rules, and different exploration-exploitation strategies. We show a series of examples for (i) how the discovery campaigns for finding materials satisfying a relative stability objective can be simulated to design new agents, and (ii) how those agents can be deployed in real discovery campaigns to control experiments run externally, such as the cloud-based density functional theory simulations in this work. In a sample set of 16 campaigns covering a range of binary and ternary chemistries including metal oxides, phosphides, sulfides and alloys, this autonomous platform found 383 new stable or nearly stable materials with no intervention by the researchers. Read More

Citation: Montoya, Joseph H., Kirsten Winther, Raul A. Flores, Thomas Bligaard, Jens Strabo Hummelshøj, and Muratahan Aykol. "Autonomous intelligent agents for accelerated materials discovery." (2020). ChemRxiv. 

TRI Authors: German Ros

All Authors: Michael R Anderson, Michael Cafarella, Thomas F Wenisch, German Ros

Querying the content of images, video, and other non-textual data sources requires expensive content extraction methods. Modern extraction techniques are based on deep convolutional neural networks (CNNs) and can classify objects within images with astounding accuracy. Unfortunately, these methods are slow, needing several milliseconds per image using modern GPUs. The cost of content-based queries over a huge video corpus is prohibitive.  Read More

Citation: Anderson, Michael R., Michael Cafarella, Thomas F. Wenisch, and German Ros. "Predicate optimization for a visual analytics database." SysML 2018

TRI Author: Brian D. Storey

All Authors: Hongbo Zhao, Brian D. Storey, Richard D. Braatz, and Martin Z. Bazant

Using a framework of partial differential equation-constrained optimization, we demonstrate that multiple constitutive relations can be extracted simultaneously from a small set of images of pattern formation. Examples include state-dependent properties in phase-field models, such as the diffusivity, kinetic prefactor, free energy, and direct correlation function, given only the general form of the Cahn-Hilliard equation, Allen-Cahn equation, or dynamical density functional theory (phase-field crystal model). Constraints can be added based on physical arguments to accelerate convergence and avoid spurious results. Reconstruction of the free energy functional, which contains nonlinear dependence on the state variable and differential or convolutional operators, opens the possibility of learning nonequilibrium thermodynamics from only a few snapshots of the dynamics. Read More

Citation: Zhao, Hongbo, Brian D. Storey, Richard D. Braatz, and Martin Z. Bazant. "Learning the Physics of Pattern Formation from Images." Physical Review Letters 124, no. 6 (2020): 060201.

TRI Author: Montoya, J. H.

All Authors: Mrdjenovich, D., Horton, M. K., Montoya, J. H., Legaspi, C. M., Dwaraknath, S., Tshitoyan, V., Jain A., Persson, K. A

Data-driven materials science is bolstered by the recent growth of online materials databases. However, the current informatics infrastructure has yet to unlock the full knowledge available within existing datasets or to explore connections between different materials science domains. Here, we present a streamlined system for codifying and connecting materials properties in an open-source Python framework: propnet. We demonstrate the capability of this framework to augment existing datasets of materials properties: by consecutively applying a network of physical relationships to calculate related information, propnet connects disparate domain knowledge. Beyond an immediate increase in available information, the results allow for the examination of correlations between sets of properties and guide the design of multifunctional materials. By emphasizing code extensibility and simplicity, we offer this software to the materials science community for general application to any experimental or computationally derived materials database.  Read More

Citations: Mrdjenovich, David, Matthew K. Horton, Joseph H. Montoya, Christian M. Legaspi, Shyam Dwaraknath, Vahe Tshitoyan, Anubhav Jain, and Kristin A. Persson. "propnet: A Knowledge Graph for Materials Science." Matter (2020).

TRI Authors: Brian Rohr, Joseph Montoya, Santosh Suram, Linda Hung

All Authors: Steven Torrisi, Matthew Carbone, Brian Rohr, Joseph Montoya, Yang Ha, Junko Yano, Santosh Suram, Linda Hung

X-ray absorption spectroscopy (XAS) produces a wealth of information about the local structure of materials, but interpretation of spectra often relies on easily accessible trends and prior assumptions about the structure. Recently, researchers have demonstrated that machine learning models can automate this process to predict the coordinating environments of absorbing atoms from their XAS spectra. However, machine learning models are often difficult to interpret, making it challenging to determine when they are valid and whether they are consistent with physical theories. In this work, we present three main advances to the data-driven analysis of XAS spectra: we demonstrate the efficacy of random forests in solving two new property determination tasks (predicting Bader charge and mean nearest neighbor distance), we show that multiscale featurization can elucidate the regions and trends in spectra that encode various local properties, and we address the effect of normalization on model interpretability. The multiscale featurization transforms the spectrum into a vector of polynomial-fit features, and is contrasted with the commonly-used "pointwise" featurization that directly uses the entire spectrum as input. We find that across thousands of transition metal oxide spectra, the relative importance of features describing the curvature of the spectrum can be localized to individual energy ranges, and we can separate the importance of constant, linear, quadratic, and cubic trends, as well as the white line energy. This work has the potential to assist rigorous theoretical interpretations, expedite experimental data collection, and automate analysis of XAS spectra, thus accelerating discovery of new functional materials.   Read More

Citation: Torrisi, Steven, Matthew Carbone, Brian Rohr, Joseph H. Montoya, Yang Ha, Junko Yano, Santosh Suram, and Linda Hung. "Random Forest Machine Learning Models for Interpretable X-Ray Absorption Near-Edge Structure Spectrum-Property Relationships." chemRxiv preprint (2020). doi:10.26434/chemrxiv.11873691.v1

TRI Authors: Muratahan Aykol, Santosh K. Suram* All Authors: Brian Rohr, Helge S. Stein, Dan Guevarra, Yu Wang, Joel A. Haber, Muratahan Aykol, Santosh K. Suram* and John M. Gregoire* 

Sequential learning (SL) strategies, i.e. iteratively updating a machine learning model to guide experiments, have been proposed to significantly accelerate materials discovery and research. Applications on computational datasets and a handful of optimization experiments have demonstrated the promise of SL, motivating a quantitative evaluation of its ability to accelerate materials discovery, specifically in the case of physical experiments. The benchmarking effort in the present work quantifies the performance of SL algorithms with respect to a breadth of research goals: discovery of any “good” material, discovery of all “good” materials, and discovery of a model that accurately predicts the performance of new materials. To benchmark the effectiveness of different machine learning models against these goals, we use datasets in which the performance of all materials in the search space is known from high-throughput synthesis and electrochemistry experiments. Each dataset contains all pseudo-quaternary metal oxide combinations from a set of six elements (chemical space), the performance metric chosen is the electrocatalytic activity (overpotential) for the oxygen evolution reaction (OER). A diverse set of SL schemes is tested on four chemical spaces, each containing 2121 catalysts. The presented work suggests that research can be accelerated by up to a factor of 20 compared to random acquisition in specific scenarios. The results also show that certain choices of SL models are ill-suited for a given research goal resulting in substantial deceleration compared to random acquisition methods. The results provide quantitative guidance on how to tune an SL strategy for a given research goal and demonstrate the need for a new generation of materials-aware SL algorithms to further accelerate materials discovery. Read More Citation: Rohr, Brian, Helge S. Stein, Dan Guevarra, Yu Wang, Joel A. Haber, Muratahan Aykol, Santosh K. Suram, and John M. Gregoire. "Benchmarking the acceleration of materials discovery by sequential learning." Chemical Science 11, no. 10 (2020): 2696-2706.

TRI Authors: Alejandro Castro, Ante Qu (intern), Naveen Kuppuswamy, Alex Alspach, Michael Sherman

All Authors: Alejandro Castro, Ante Qu, Naveen Kuppuswamy, Alex Alspach, Michael Sherman

Multibody simulation with frictional contact has been a challenging subject of research for the past thirty years. Rigidbody assumptions are commonly used to approximate the physics of contact, and together with Coulomb friction, lead to challengingto-solve nonlinear complementarity problems (NCP). On the other hand, robot grippers often introduce significant compliance. Compliant contact, combined with regularized friction, can be modeled entirely with ODEs, avoiding NCP solves. Unfortunately, regularized friction introduces high-frequency stiff dynamics and even implicit methods struggle with these systems, especially during slip-stick transitions. To improve the performance of implicit integration for these systems we introduce a Transition-Aware Line Search (TALS), which greatly improves the convergence of the Newton-Raphson iterations performed by implicit integrators. We find that TALS works best with semi-implicit integration, but that the explicit treatment of normal compliance can be problematic. To address this, we develop a Transition-Aware Modified SemiImplicit (TAMSI) integrator that has similar computational cost to semi-implicit methods but implicitly couples compliant contact forces, leading to a more robust method. We evaluate the robustness, accuracy and performance of TAMSI and demonstrate our approach alongside relevant sim-to-real manipulation tasks. Read More

Citation: Castro, Alejandro M., Ante Qu, Naveen Kuppuswamy, Alex Alspach, and Michael Sherman. "A Transition-Aware Method for the Simulation of Compliant Contact with Regularized Friction." IEEE Robotics and Automation Letters 5, no. 2 (2020): 1859-1866.

Battery evaluation and early prediction software package (BEEP) provides an open-source Python-based framework for the management and processing of high-throughput battery cycling data-streams. BEEPs features include file-system based organization of raw cycling data and metadata received from cell testing equipment, validation protocols that ensure the integrity of such data, parsing and structuring of data into Python-objects ready for analytics, featurization of structured cycling data to serve as input for machine-learning, and end-to-end examples that use processed data for anomaly detection and featurized data to train early-prediction models for cycle life. BEEP is developed in response to the software and expertise gap between cell-level battery testing and data-driven battery development. READ MORE

TRI Authors: Aykol, Muratahan, Jens S. Hummelshøj, Abraham Anapolsky, Chirranjeevi Balaji Gopal, Patrick Herring, Linda Hung, Ha-Kyung Kwon, Joseph H. Montoya, Daniel Schweigert, Santosh K. Suram, and Brian D. Storey

All Authors: Aykol, Muratahan, Jens S. Hummelshøj, Abraham Anapolsky, Koutarou Aoyagi, Martin Z. Bazant, Thomas Bligaard, Richard D. Braatz, Scott Broderick, Daniel Cogswell, John Dagdelen, Walter Drisdell, Edwin Garcia, Krishna Garipati, Vikram Gavini, William E. Gent, Livia Giordano, Carla P. Gomes, Rafael Gomez-Bombarelli, Chirranjeevi Balaji Gopal, John M. Greguire, Jeffrey C. Grossman, Patrick Herring, Linda Hung, Thomas F. Jaramillo, Laurie King, Ha-Kyung Kwon, Ryosuke Maekawa, Andrew M. Minor, Joseph H. Montoya, Tim Mueller, Colin Ophus, Krishna Rajan, Rampi Ramprasad, Brian Rohr, Daniel Schweigert, Yang Shao-Horn, Yoshinori Suga, Santosh K. Suram, Venkatasubramanian Viswanathan, Jay F Whitacre, Adam P. Willard, Olga Wodo, Chris Wolverton, and Brian D. Storey Read More

Citation:  Aykol, Muratahan, Jens S. Hummelshøj, Abraham Anapolsky, Koutarou Aoyagi, Martin Z. Bazant, Thomas Bligaard, Richard D. Braatz et al. "The Materials Research Platform: Defining the Requirements from User Stories." Matter 1, no. 6 (2019): 1433-1438.