DMSE Master's Thesis Presentation - William H. Harris
Thursday, August 06, 2020 at 12:00am to 1:00pmVirtual Event
DMSE Master’s Thesis Presentation
Machine Learning Transferable Physics-Based Force Fields using Graph Convolutional Neural Networks
William H. Harris
Thursday, August 6, 2020
12:00 am – 1:00 pm EDT
Contact firstname.lastname@example.org for guest link.
Molecular dynamics and Monte Carlo methods allow the properties of a system to be determined from its potential energy surface (PES). In the domain of crystalline materials, the PES is needed for electronic structure calculations, critical for modeling semiconductors, optical, and energy-storage materials. While first principles techniques can be used to obtain the PES to high accuracy, their computational complexity limits applications to small systems and short timescales. In practice, the PES must be approximated using a computationally cheaper functional form. Classical force field (CFF) approaches simply define the PES as a sum over independent energy contributions. Commonly included terms include bonded (pair, angle, dihedral, etc.) and non-bonded (van der Waals, Coulomb, etc.) interactions, while more recent CFFs model polarizability, reactivity, and other higher-order interactions. Simple, physically-justified functional forms are often implemented for each energy type, but this choice – and the choice of which energy terms to include in the first place – is arbitrary and often hand-tuned on a per-system basis, severely limiting PES transferability. This flexibility has complicated the quest for a universal CFF. The simplest usable CFFs are tailored to specific classes of molecules and have few parameters, so that they can be optimally parameterized using a small amount of data; however, they suffer low transferability. Highly-parameterized neural network potentials can yield predictions that are extremely accurate for the entire training set; however, they suffer over-fitting and cannot interpolate. We explore the trade-offs between complexity and generalizability in fitting CFFs; focus on simple, computationally fast functions that enforce physics-based regularization and transferability; use available QM big-data to densely sample chemical space; and utilize high performance computing resources to expand training data and the number of CFF implementations that can be addressed. A universal, fast CFF would open the door to high-throughput virtual materials screening in the pursuit of novel materials with tailored properties.
Rafael Gomez-Bombarelli, Toyota Assistant Professor in Materials Processing, Materials Science and Engineering, Massachusetts Institute of Technology