-
Multimodal machine learning for materials science: composition-structure bimodal learning for experimentally measured properties
Authors:
Sheng Gong,
Shuo Wang,
Taishan Zhu,
Yang Shao-Horn,
Jeffrey C. Grossman
Abstract:
The widespread application of multimodal machine learning models like GPT-4 has revolutionized various research fields including computer vision and natural language processing. However, its implementation in materials informatics remains underexplored, despite the presence of materials data across diverse modalities, such as composition and structure. The effectiveness of machine learning models…
▽ More
The widespread application of multimodal machine learning models like GPT-4 has revolutionized various research fields including computer vision and natural language processing. However, its implementation in materials informatics remains underexplored, despite the presence of materials data across diverse modalities, such as composition and structure. The effectiveness of machine learning models trained on large calculated datasets depends on the accuracy of calculations, while experimental datasets often have limited data availability and incomplete information. This paper introduces a novel approach to multimodal machine learning in materials science via composition-structure bimodal learning. The proposed COmposition-Structure Bimodal Network (COSNet) is designed to enhance learning and predictions of experimentally measured materials properties that have incomplete structure information. Bimodal learning significantly reduces prediction errors across distinct materials properties including Li conductivity in solid electrolyte, band gap, refractive index, dielectric constant, energy, and magnetic moment, surpassing composition-only learning methods. Furthermore, we identified that data augmentation based on modal availability plays a pivotal role in the success of bimodal learning.
△ Less
Submitted 3 August, 2023;
originally announced September 2023.
-
The Disparate Impacts of College Admissions Policies on Asian American Applicants
Authors:
Joshua Grossman,
Sabina Tomkins,
Lindsay Page,
Sharad Goel
Abstract:
There is debate over whether Asian American students are admitted to selective colleges and universities at lower rates than white students with similar academic qualifications. However, there have been few empirical investigations of this issue, in large part due to a dearth of data. Here we present the results from analyzing 685,709 applications from Asian American and white students to a subset…
▽ More
There is debate over whether Asian American students are admitted to selective colleges and universities at lower rates than white students with similar academic qualifications. However, there have been few empirical investigations of this issue, in large part due to a dearth of data. Here we present the results from analyzing 685,709 applications from Asian American and white students to a subset of selective U.S. institutions over five application cycles, beginning with the 2015-2016 cycle. The dataset does not include admissions decisions, and so we construct a proxy based in part on enrollment choices. Based on this proxy, we estimate the odds that Asian American applicants were admitted to at least one of the schools we consider were 28% lower than the odds for white students with similar test scores, grade-point averages, and extracurricular activities. The gap was particularly pronounced for students of South Asian descent (49% lower odds). We trace this pattern in part to two factors. First, many selective colleges openly give preference to the children of alumni, and we find that white applicants were substantially more likely to have such legacy status than Asian applicants, especially South Asian applicants. Second, after adjusting for observed student characteristics, the institutions we consider appear less likely to admit students from geographic regions with relatively high shares of applicants who are Asian. We hope these results inform ongoing discussions on the equity of college admissions policies.
△ Less
Submitted 3 August, 2023;
originally announced August 2023.
-
A cloud platform for automating and sharing analysis of raw simulation data from high throughput polymer molecular dynamics simulations
Authors:
Tian Xie,
Ha-Kyung Kwon,
Daniel Schweigert,
Sheng Gong,
Arthur France-Lanord,
Arash Khajeh,
Emily Crabb,
Michael Puzon,
Chris Fajardo,
Will Powelson,
Yang Shao-Horn,
Jeffrey C. Grossman
Abstract:
Open material databases storing hundreds of thousands of material structures and their corresponding properties have become the cornerstone of modern computational materials science. Yet, the raw outputs of the simulations, such as the trajectories from molecular dynamics simulations and charge densities from density functional theory calculations, are generally not shared due to their huge size.…
▽ More
Open material databases storing hundreds of thousands of material structures and their corresponding properties have become the cornerstone of modern computational materials science. Yet, the raw outputs of the simulations, such as the trajectories from molecular dynamics simulations and charge densities from density functional theory calculations, are generally not shared due to their huge size. In this work, we describe a cloud-based platform to facilitate the sharing of raw data and enable the fast post-processing in the cloud to extract new properties defined by the user. As an initial demonstration, our database currently includes 6286 molecular dynamics trajectories for amorphous polymer electrolytes and 5.7 terabytes of data. We create a public analysis library at https://github.com/TRI-AMDD/htp_md to extract multiple properties from the raw data, using both expert designed functions and machine learning models. The analysis is run automatically with computation in the cloud, and results then populate a database that can be accessed publicly. Our platform encourages users to contribute both new trajectory data and analysis functions via public interfaces. Newly analyzed properties will be incorporated into the database. Finally, we create a front-end user interface at https://www.htpmd.matr.io for browsing and visualization of our data. We envision the platform to be a new way of sharing raw data and new insights for the computational materials science community.
△ Less
Submitted 2 August, 2022;
originally announced August 2022.
-
The Specialized High-Performance Network on Anton 3
Authors:
Keun Sup Shim,
Brian Greskamp,
Brian Towles,
Bruce Edwards,
J. P. Grossman,
David E. Shaw
Abstract:
Molecular dynamics (MD) simulation, a computationally intensive method that provides invaluable insights into the behavior of biomolecules, typically requires large-scale parallelization. Implementation of fast parallel MD simulation demands both high bandwidth and low latency for inter-node communication, but in current semiconductor technology, neither of these properties is scaling as quickly a…
▽ More
Molecular dynamics (MD) simulation, a computationally intensive method that provides invaluable insights into the behavior of biomolecules, typically requires large-scale parallelization. Implementation of fast parallel MD simulation demands both high bandwidth and low latency for inter-node communication, but in current semiconductor technology, neither of these properties is scaling as quickly as intra-node computational capacity. This disparity in scaling necessitates architectural innovations to maximize the utilization of computational units. For Anton 3, the latest in a family of highly successful special-purpose supercomputers designed for MD simulations, we thus designed and built a completely new specialized network as part of our ASIC. Tightly integrating this network with specialized computation pipelines enables Anton 3 to perform simulations orders of magnitude faster than any general-purpose supercomputer, and to outperform its predecessor, Anton 2 (the state of the art prior to Anton 3), by an order of magnitude. In this paper, we present the three key features of the network that contribute to the high performance of Anton 3. First, through architectural optimizations, the network achieves very low end-to-end inter-node communication latency for fine-grained messages, allowing for better overlap of computation and communication. Second, novel application-specific compression techniques reduce the size of most messages sent between nodes, thereby increasing effective inter-node bandwidth. Lastly, a new hardware synchronization primitive, called a network fence, supports fast fine-grained synchronization tailored to the data flow within a parallel MD application. These application-driven specializations to the network are critical for Anton 3's MD simulation performance advantage over all other machines.
△ Less
Submitted 20 January, 2022;
originally announced January 2022.
-
Accelerating amorphous polymer electrolyte screening by learning to reduce errors in molecular dynamics simulated properties
Authors:
Tian Xie,
Arthur France-Lanord,
Yanming Wang,
Jeffrey Lopez,
Michael Austin Stolberg,
Megan Hill,
Graham Michael Leverick,
Rafael Gomez-Bombarelli,
Jeremiah A. Johnson,
Yang Shao-Horn,
Jeffrey C. Grossman
Abstract:
Polymer electrolytes are promising candidates for the next generation lithium-ion battery technology. Large scale screening of polymer electrolytes is hindered by the significant cost of molecular dynamics (MD) simulation in amorphous systems: the amorphous structure of polymers requires multiple, repeated sampling to reduce noise and the slow relaxation requires long simulation time for convergen…
▽ More
Polymer electrolytes are promising candidates for the next generation lithium-ion battery technology. Large scale screening of polymer electrolytes is hindered by the significant cost of molecular dynamics (MD) simulation in amorphous systems: the amorphous structure of polymers requires multiple, repeated sampling to reduce noise and the slow relaxation requires long simulation time for convergence. Here, we accelerate the screening with a multi-task graph neural network that learns from a large amount of noisy, unconverged, short MD data and a small number of converged, long MD data. We achieve accurate predictions of 4 different converged properties and screen a space of 6247 polymers that is orders of magnitude larger than previous computational studies. Further, we extract several design principles for polymer electrolytes and provide an open dataset for the community. Our approach could be applicable to a broad class of material discovery problems that involve the simulation of complex, amorphous materials.
△ Less
Submitted 15 March, 2022; v1 submitted 13 January, 2021;
originally announced January 2021.
-
Graph Dynamical Networks for Unsupervised Learning of Atomic Scale Dynamics in Materials
Authors:
Tian Xie,
Arthur France-Lanord,
Yanming Wang,
Yang Shao-Horn,
Jeffrey C. Grossman
Abstract:
Understanding the dynamical processes that govern the performance of functional materials is essential for the design of next generation materials to tackle global energy and environmental challenges. Many of these processes involve the dynamics of individual atoms or small molecules in condensed phases, e.g. lithium ions in electrolytes, water molecules in membranes, molten atoms at interfaces, e…
▽ More
Understanding the dynamical processes that govern the performance of functional materials is essential for the design of next generation materials to tackle global energy and environmental challenges. Many of these processes involve the dynamics of individual atoms or small molecules in condensed phases, e.g. lithium ions in electrolytes, water molecules in membranes, molten atoms at interfaces, etc., which are difficult to understand due to the complexity of local environments. In this work, we develop graph dynamical networks, an unsupervised learning approach for understanding atomic scale dynamics in arbitrary phases and environments from molecular dynamics simulations. We show that important dynamical information can be learned for various multi-component amorphous material systems, which is difficult to obtain otherwise. With the large amounts of molecular dynamics data generated everyday in nearly every aspect of materials design, this approach provides a broadly useful, automated tool to understand atomic scale dynamics in material systems.
△ Less
Submitted 22 May, 2019; v1 submitted 18 February, 2019;
originally announced February 2019.
-
Hierarchical Visualization of Materials Space with Graph Convolutional Neural Networks
Authors:
Tian Xie,
Jeffrey C. Grossman
Abstract:
The combination of high throughput computation and machine learning has led to a new paradigm in materials design by allowing for the direct screening of vast portions of structural, chemical, and property space. The use of these powerful techniques leads to the generation of enormous amounts of data, which in turn calls for new techniques to efficiently explore and visualize the materials space t…
▽ More
The combination of high throughput computation and machine learning has led to a new paradigm in materials design by allowing for the direct screening of vast portions of structural, chemical, and property space. The use of these powerful techniques leads to the generation of enormous amounts of data, which in turn calls for new techniques to efficiently explore and visualize the materials space to help identify underlying patterns. In this work, we develop a unified framework to hierarchically visualize the compositional and structural similarities between materials in an arbitrary material space with representations learned from different layers of graph convolutional neural networks. We demonstrate the potential for such a visualization approach by showing that patterns emerge automatically that reflect similarities at different scales in three representative classes of materials: perovskites, elemental boron, and general inorganic crystals, covering material spaces of different compositions, structures, and both. For perovskites, elemental similarities are learned that reflects multiple aspects of atom properties. For elemental boron, structural motifs emerge automatically showing characteristic boron local environments. For inorganic crystals, the similarity and stability of local coordination environments are shown combining different center and neighbor atoms. The method could help transition to a data-centered exploration of materials space in automated materials design.
△ Less
Submitted 18 September, 2018; v1 submitted 9 July, 2018;
originally announced July 2018.
-
Machine Learning Enabled Computational Screening of Inorganic Solid Electrolytes for Dendrite Suppression with Li Metal Anode
Authors:
Zeeshan Ahmad,
Tian Xie,
Chinmay Maheshwari,
Jeffrey C. Grossman,
Venkatasubramanian Viswanathan
Abstract:
Next generation batteries based on lithium (Li) metal anodes have been plagued by the dendritic electrodeposition of Li metal on the anode during cycling, resulting in short circuit and capacity loss. Suppression of dendritic growth through the use of solid electrolytes has emerged as one of the most promising strategies for enabling the use of Li metal anodes. We perform a computational screening…
▽ More
Next generation batteries based on lithium (Li) metal anodes have been plagued by the dendritic electrodeposition of Li metal on the anode during cycling, resulting in short circuit and capacity loss. Suppression of dendritic growth through the use of solid electrolytes has emerged as one of the most promising strategies for enabling the use of Li metal anodes. We perform a computational screening of over 12,000 inorganic solids based on their ability to suppress dendrite initiation in contact with Li metal anode. Properties for mechanically isotropic and anisotropic interfaces that can be used in stability criteria for determining the propensity of dendrite initiation are usually obtained from computationally expensive first-principles methods. In order to obtain a large dataset for screening, we use machine learning models to predict the mechanical properties of several new solid electrolytes. We train a convolutional neural network on the shear and bulk moduli purely on structural features of the material. We use AdaBoost, Lasso and Bayesian ridge regression to train the elastic constants, where the choice of the model depended on the size of the training data and the noise that it can handle. Our models give us direct interpretability by revealing the dominant structural features affecting the elastic constants. The stiffness is found to increase with a decrease in volume per atom, increase in minimum anion-anion separation, and increase in sublattice (all but Li) packing fraction. Cross-validation/test performance suggests our models generalize well. We predict over 20 mechanically anisotropic interfaces between Li metal and 6 solid electrolytes which can be used to suppress dendrite growth. Our screened candidates are generally soft and highly anisotropic, and present opportunities for simultaneously obtaining dendrite suppression and high ionic conductivity in solid electrolytes.
△ Less
Submitted 12 April, 2018;
originally announced April 2018.
-
Beyond electronics, beyond optics: single circuit parallel computing with phonons
Authors:
Sophia Sklan,
Jeffrey C. Grossman
Abstract:
Phononic computing -- the use of (typically thermal) vibrations for information processing -- is a nascent technology; its capabilities are still being discovered. We analyze an alternative form of phononic computing inspired by optical, rather than electronic, computing. Using the acoustic Faraday effect, we design a phonon gyrator and thereby a means of performing computation through the manipul…
▽ More
Phononic computing -- the use of (typically thermal) vibrations for information processing -- is a nascent technology; its capabilities are still being discovered. We analyze an alternative form of phononic computing inspired by optical, rather than electronic, computing. Using the acoustic Faraday effect, we design a phonon gyrator and thereby a means of performing computation through the manipulation of polarization in transverse phonon currents. Moreover, we establish that our gyrators act as generalized transistors and can construct digital logic gates. Exploiting the wave nature of phonons and the similarity of our logic gates, we demonstrate parallel computation within a single circuit, an effect presently unique to phonons. Finally, a generic method of designing these parallel circuits is introduced and used to analyze the feasibility of magneto-acoustic materials in realizing these circuits.
△ Less
Submitted 13 January, 2013;
originally announced January 2013.
