ProgramMaster Logo
Conference Tools for 2018 TMS Annual Meeting & Exhibition
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools
About this Abstract
Meeting 2018 TMS Annual Meeting & Exhibition
Symposium Computational Materials Discovery and Optimization
Presentation Title Quantum-accurate Force Fields from Machine Learning of Large Materials Data
Author(s) Shyue Ping Ong, Chi Chen, Zhi Deng, Richard Tran
On-Site Speaker (Planned) Shyue Ping Ong
Abstract Scope In this talk, I will discuss how the rigorous application of machine learning techniques on large materials data sets may be used to develop efficient, quantum-accurate force-fields for elemental metals (e.g., Mo, Li) as well as complex multi-species compounds. We will outline a systematic approach to structural selection based on principal component analysis, as well as a novel differential evolution algorithm for optimizing the hyperparameters in the model fitting so that both the model error and the property prediction error can be simultaneously lowered. We will demonstrate that this force field can successfully predict a broad range of properties, such as energies, forces, elastic constants, melting point, phonon spectra, surface energies, etc. with accuracy close to that of DFT computations, outperforming traditional force fields based on the embedded atom (EAM) and modified embedded atom methods (MEAM). We expect that these techniques will find broad application in large-scale, long-time scale simulations.
Proceedings Inclusion? Planned: Supplemental Proceedings volume

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

A Combined Experimental-computational Approach to Determining Nanoscale Structures
A Materials-informatics Approach for Finding New Hard-magnetic Phases
Computational Design of Fatigue-resistant NiTi-based Shape Memory Alloys
Computational Screening of Novel Two-dimensional Topological Insulators and Layer-dependent Properties
Data-driven Discovery of Photocathodes for CO2 Reduction
Design Concepts of Optimized MRI Magnet by COMSOL Multiphysics Simulation
Determination of Thermal Transport in Solids and Liquids by Non-equilibrium Molecular Dynamics Simulations
Dual Band Metamaterial Perfect Absorber Based on Mie Resonances
Economic Analysis of National Needs for Technology Infrastructure to Support the Materials Genome Initiative
Fabricating Optimized Crystallographic Textures through Heterogeneous Templated Grain Growth
First-principles Calculations on the Multiferroic Properties of Two-dimensional Oxides
First Principle Prediction of Magnetic Topological Phase in Thin Films of Bi2XY4 (X = Mn, Cr; Y = Se, Te)
High-throughput Investigation of the Electronic Properties of 2D and Bulk Materials in the MaterialsWeb Database
Holistic Computational Structure Screening of More than 12 000 Candidates for Solid Lithium-ion Conductor Materials
Improving the Ductility of Boron Carbide from Computational Design
Learning Grain Boundary Properties from Macroscopic and Microscopic Structural Descriptors
Light-metal Complex Hydrides: Computational Structure Prediction and Interaction with Functionalized Nanoporous Hosts
Machine Learning for Materials
Machine Learning for Prediction of Electronic Structures of Multi-component Alloys
Minimal Addition of Cerium for Stability of Critical Phases in Hard Magnetic AlNiCo Alloys: Combined Machine Learning and CALPHAD
Molecular Crystal Structure Prediction with Gator and Genarris
Predicting Ferroelectric Properties from Microstructures with Deep Learning
Quantum-accurate Force Fields from Machine Learning of Large Materials Data
Reentrant Melting of Sodium, Magnesium and Aluminum and Possible Universal Trend
Search for Rare-Earth Free Permanent Magnets in Fe and Co Based Compounds by Adaptive Genetic Algorithm
Software Tools for High-throughput Materials Data Generation and Data Mining
Structure-property Linkages for Porous Membranes Using the Materials Knowledge Systems Framework
Tailoring Properties in Multi-component Alloys through Heuristic Optimization
The Use of Cluster Expansions to Predict the Structure and Properties of Catalysts

Questions about ProgramMaster? Contact programming@programmaster.org