Bulletin of the American Physical Society
APS March Meeting 2023
Las Vegas, Nevada (March 510)
Virtual (March 2022); Time Zone: Pacific Time
Session Q53: AI and Statistical/Thermal PhysicsFocus

Hide Abstracts 
Sponsoring Units: GDS Chair: Caroline Desgranges, University of Massachusetts Lowell Room: Room 307 
Wednesday, March 8, 2023 3:00PM  3:36PM 
Q53.00001: Selfassembly of electronic materials and the power of machine learning Invited Speaker: Paulette Clancy There are many problems at the forefront of materials chemistry whose solution is stymied by its inherent complexity. Such problems are characterized by a rich landscape of parameters and processing variables that is combinatorially too large for either an experimental or a computational approach to solve through an exhaustive search. In such cases, the usual approach is an Edisonian trialanderror approach, which inevitably leaves areas of parameter space unexplored. The problems that we have explored are also characterized by a scarcity of data, since the data are expensive to acquire, both experimentally and computationally. This makes it an ideal candidate to solve using a Bayesian optimization (BayesOpt) approach. 
Wednesday, March 8, 2023 3:36PM  3:48PM 
Q53.00002: How deep neural networks learn thermal phase transitions Julian Arnold, Frank Schäfer Machinelearning methods have successfully been used to identify phase transitions from data. Neural network (NN)based approaches are particularly appealing due to the ability of NNs to learn arbitrary functions. However, the larger an NN, the more computational resources are needed to train it, and the more difficult it is to understand its decision making. We derive analytical expressions for the optimal output of three popular NNbased methods for detecting phase transitions which rely on solving classification and regression tasks using supervised learning at their core [1]. This result corresponds to the output when using optimal predictive models, such as sufficiently large NNs after ideal training. Our analysis reveals that highcapacity neural networks ultimately gauge changes in the energy distributions as a function of temperature when used to detect thermal phase transitions. In contrast, lowcapacity neural networks seem to learn order parameters, i.e., recognize prevalent patterns or orderings. Our theoretical findings are supported by analyzing data from numerical simulations of classical spin systems. 
Wednesday, March 8, 2023 3:48PM  4:00PM 
Q53.00003: Machine learning phases of matter: Scalability and limitations Zhongzheng Tian, Sheng Zhang, GiaWei Chern We present a scalable machine learning (ML) framework for distinguishing phases and identifying phase transitions in manybody systems. Scalability and transferability are central to the unprecedented computational efficiency of ML methods. In general, linearscaling computation can be achieved through the divide and conquer approach, and the locality of physical properties is key to partitioning the system into subdomains that can be solved individually. Our proposed ML framework for phase classification is similar in spirit to the BehleParrinello approach widely employed in MLbased quantum molecular dynamics simulations. We discuss the limitations of this approach and the important role of the correlation length. The prototypical twodimensional Ising model is used to demonstrate the proposed framework. Implications for practical applications are also discussed. 
Wednesday, March 8, 2023 4:00PM  4:12PM 
Q53.00004: Learning Together: Training Interatomic Potentials to Multiple Datasets Alice Allen The development of machine learning interatomic potentials has led to an abundance of datasets containing quantum mechanical calculations for molecular and material systems. However, using the information from different datasets together remains a challenge due to the varying levels of theory employed. In this talk, we show that techniques can be used to fit an interatomic potential to multiple organic molecule datasets and that this yields ML potentials with improved accuracies for a variety of tasks. 
Wednesday, March 8, 2023 4:12PM  4:24PM 
Q53.00005: Local force field of thermally displaced atoms in unstable bcc iron from machine learning Adrian De la Rocha Galán, Valeria I Arteaga Muniz, Blaise A Ayirizia, Sofia G Gomez, Ramon J Ravelo, Wibe A de Jong, Jorge A Munoz A dataset of energy versus atomic thermal displacements was created from density functional theory molecular dynamics simulations of nonspinpolarized bodycentered cubic iron at pressures centered at 7 GPa and temperatures centered at 1700K, and Gaussian process regression was used to make predictions based on the similarity between atomic displacements as determined by a graph kernel. Force versus displacement relationships were computed at randomly selected timesteps of the simulations for all atoms in the supercell in the directions of their first, second, and third nearestneighbors. The atoms experience a generally restorative force in the direction of their first and thirdnearest neighbors, but are unstable in the direction of their secondnearest neighbors. The predicted dynamics are consistent with a martensitic phase transition to facecentered cubic, which is the thermodynamically stable phase of iron at the investigated temperature and pressure conditions. 
Wednesday, March 8, 2023 4:24PM  4:36PM 
Q53.00006: CHGNet: Pretrained Neural Network Potential for Fast and Accurate Chargeconstrained Molecular Dynamics Bowen Deng, Peichen Zhong, Gerbrand Ceder Molecular dynamics (MD) simulation coupled on systems with complex electron interactions remains one of the biggest challenges for atomistic modeling. While classical force fields often fail to describe the electronic coupling with ionic rearrangements, the more accurate spinpolarized ab initio molecular dynamics (AIMD) suffer from the computational complexity that prevents longtime and largescale simulation, which are essential to study ion migrations and phase transformations. 
Wednesday, March 8, 2023 4:36PM  4:48PM 
Q53.00007: Neuralnetworkbased interatomic potential: A case study on lithium Naman Katyal Advancements in neuralnetworkbased force fields have led to the predictions of materials for applications in widespread applications. In this work, we will show a general scheme that can be used to develop a neuralnetworkbased interatomic potential using our inhouse developed python atomcentered machine learning force field package (PyAMFF) with GPU capabilities. Using an example of lithium, we will show a force field can be developed using neural networks: (a) Data Collection step: atomic positions, energies, and forces from density functional theory (DFT) calculations for different lithium systems; (b) Fingerprint Selection step: an automated BehlerParrinello representation selection for a dataset using radial and angular distribution function; (c) Training Dataset Generation step: selection criteria in fingerprint space to reduce the size of DFT dataset for neural network training; (d) Model Training step: analyzing the effect of neural network size and fingerprint selection on model accuracy; (e) Model Performance step: rigorous testing of neuralnetworkbased force field on rare event searches using Adaptive Kinetic Monte Carlo and global optimization of lithium clusters using basin hopping. This force field will help in answering questions related to kinetics of lithium deposition on lithium metal surfaces at experimental timescales for battery applications. 
Wednesday, March 8, 2023 4:48PM  5:00PM 
Q53.00008: Onthefly Machine LearningAccelerated Geometry Optimization: Theoretical Screening of a Single Atom Alloy for CO_{2} Electroreduction Reaction Jiyoung Lee Forcefields based upon machine learning (ML) methods are being adopted for the design of catalysts due to their high efficiency and firstprinciples level accuracy. Such efficiency and accuracy are important particularly for highthroughput computations, which are key to the computational design of new catalysts. Among various types of catalysts, singleatom alloy (SAA) catalysts, which substitute a single atom from the surface for a different type of atom, have exhibited outstanding selectivity and activity due to their unique geometry. There have been a number of theoretical studies to find optimal combinations of SAA for catalytic reactions but screening all possible combinations remains intractable due to the computational expense. Here, we show a mathematical modeling package of training potential energy surfaces using artificial neural networks and applying the machinelearned models to accelerate geometry optimization process. Onthefly generated machine learning forcefields are used as the basis for a separate optimization to reach a local minima or saddle point on the surrogate PES, which requires less computational effort than the same steps using DFT. Iteration between MLoptimization and DFT calculations will reduce the overall number of DFT force calls required for optimizations, while retaining DFT accuracy for highthroughput screening. 
Wednesday, March 8, 2023 5:00PM  5:12PM 
Q53.00009: Investigating the influence of local composition on properties in complex alloys using machine learned interatomic potentials Megan J McCarthy, Jacob Startt, Remi Dingreville, Aidan P Thompson, Mitchell A Wood Complex concentrated alloys (CCAs) contain high concentrations of three or more metallic elements which mix exceptionally well, down to the atomic scale. In contrast to conventional alloy properties, CCA properties are highly sensitive to local chemistry and the specific scale of property measurement. For a multitude of reasons, molecular dynamics (MD) stands out as uniquely suited to probe this distinctive scalesensitivity. However, there are two major, interrelated challenges to tackle before MD can make real inroads into this area. The first is the development of transferrable interatomic potentials that can accurately model CCAs through widely varying chemical environments. Highly accurate machinelearned interatomic potentials (MLIAPs) hold enormous promise as a solution, but training and validating CCA MLIAPs is still a nascent field. The second challenge is that, even with excellent CCA MLIAPs, little progress can be made without ways to define, track, and compare the ‘local’ alloy compositions contained in a CCA sample. Relying exclusively on a traditional ‘global’ alloy composition label (such as ‘equiatomic’) to describe properties hinders understanding of the varied nanoscale chemical interactions that characterize CCAs. 
Wednesday, March 8, 2023 5:12PM  5:24PM 
Q53.00010: Nanoparticle Heterogeneous Catalysis Dynamics Simulations with Machine Learned Force Fields Cameron J Owen, Yu Xie, Jin Soo Lim, Boris Kozinsky Quantitative understanding and control of interfacial reactions between the gasphase and solid surfaces are crucial for improving numerous catalysis and energy conversion systems. Examples of these interfacial phenomena include H_{2} splitting and CO adsorption on nanoparticles, both of which are important industrial processes and lead to markedly different particle behaviors. These disparate particle responses cannot be resolved under the spatial and temporal resolutions of current experimental techniques, lending this problem to be solved by molecular dynamics simulations. We train a collection of robust, Bayesian machine learned force fields (MLFFs) using the FLARE onthefly active learning framework implemented using Gaussian Process regression. Molecular dynamics (MD) simulations are used to sample atomic configurations and density functional theory is only called upon when the Bayesian uncertainty exceeds a threshold. This workflow yields both acceleration in timetosolution and an increase in computational efficiency relative to ab initio MD. The resulting MLFFs retain first principles accuracy, are fast, and are uncertaintyaware. Following a rigorous validation scheme, through comparison of the dynamic evolution of these particles and bulk systems to available xray data (i.e., extended xray absorption fine spectra) and static benchmarks, long timescale MD simulations for freestanding metal nanoparticle systems (e.g., Pt, Au, PdAu, and CuPt) are performed. In addition to the bare particles, reaction mechanisms under gaseous exposure (e.g., H_{2} and CO) are also investigated. These MLFFs allow for the simultaneous study of atomistic mechanisms occurring on these nanoparticles under exposure to reactive atmospheres and the evolution of their structural morphologies. 
Wednesday, March 8, 2023 5:24PM  5:36PM 
Q53.00011: Melting and Phase Separation of SiC from Largescale Machine Learning Molecular Dynamics Yu Xie, Senja J Ramakers, Boris Kozinsky Incongruent melting has been reported by many experimental studies with the observation that the SiC decomposes into liquid silicon and solid carbon upon heating. However, contradictory results have also been reported that SiC melts congruently with no decomposition. Ab initio MD provides evidence of carbon clustering during the melting process but did not capture the phase separation limited by the small supercell. Empirical potentials give good approximations of the melting points but all of them have limited ability to describe other phases and thus cannot capture the decomposition in the MD simulation. In this work, we use Bayesian active learning to train a machine learning potential for SiC at a wide range of temperatures and pressures, and observe the phase decomposition process into Si and Crich phases in both the active learning and the largescale MD stages. Our results provide a microscopic validation of the incongruent melting hypothesis, resolving the disagreement among experiments of the SiC melting behavior. The occurrence of decomposition also indicates that amorphous SiC can not be produced by melting and quenching but only with irradiation, consistent with existing experimental approaches of the amorphization of SiC. 
Wednesday, March 8, 2023 5:36PM  5:48PM 
Q53.00012: ML models for partition functions: from the prediction of thermodynamic properties to the exploration of transition pathways Jerome P Delhommelle, Caroline Desgranges We discuss our recent work on the development of machine learning (ML) models for the prediction of partition functions. To this end, we carry out computationallyintensive flat histogram Monte Carlo simulations, based on a WangLandau sampling scheme, to obtain the partition function for atomic and molecular systems. The simulation results for the partition functions are then gathered in datasets that allow us to train and validate ML models. This, in turn, leads us to build artificial neural networks models for the prediction of partition functions for a broad range of conditions. We demonstrate the accuracy and reliability of the ML models by showing the ability of the MLpartition function to predict thermodynamic properties for single componentsystems, mixtures, confined fluids. Furthermore, we show that the MLpartition function can be leveraged to build reaction coordinates for the exploration of phase transition pathways. 
Wednesday, March 8, 2023 5:48PM  6:00PM 
Q53.00013: Diffeomorphisms invariance is a proxy of performance in deep neural networks Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart Understanding why deep nets can classify data in large dimensions remains a challenge. It has been proposed that they do so by becoming stable to diffeomorphisms, yet existing empirical measurements support that it is often not the case. We revisit this question by defining a maximumentropy distribution on diffeomorphisms, that allows to study typical diffeomorphisms of a given norm. We confirm that stability toward diffeomorphisms does not strongly correlate to performance on benchmark data sets of images. By contrast, we find that the stability toward diffeomorphisms relative to that of generic transformations R_{f} correlates remarkably with the test error ε_{t}. It is of order unity at initialization but decreases by several decades during training for stateoftheart architectures. For CIFAR10 and 15 known architectures, we find ε_{t} ∼ 0.2 √R_{f}, suggesting that obtaining a small R_{f} is important to achieve good performance. 
Follow Us 
Engage
Become an APS Member 
My APS
Renew Membership 
Information for 
About APSThe American Physical Society (APS) is a nonprofit membership organization working to advance the knowledge of physics. 
© 2023 American Physical Society
 All rights reserved  Terms of Use
 Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 207403844
(301) 2093200
Editorial Office
1 Research Road, Ridge, NY 119612701
(631) 5914000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 200452001
(202) 6628700