Synthetic pre-training for neural-network interatomic potentials
- URL: http://arxiv.org/abs/2307.15714v1
- Date: Mon, 24 Jul 2023 17:16:24 GMT
- Title: Synthetic pre-training for neural-network interatomic potentials
- Authors: John L. A. Gardner and Kathryn T. Baker and Volker L. Deringer
- Abstract summary: We show that synthetic atomistic data, themselves obtained at scale with an existing machine learning potential, constitute a useful pre-training task for neural-network interatomic potential models.
Once pre-trained with a large synthetic dataset, these models can be fine-tuned on a much smaller, quantum-mechanical one, improving numerical accuracy and stability in computational practice.
- Score: 0.0
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Machine learning (ML) based interatomic potentials have transformed the field
of atomistic materials modelling. However, ML potentials depend critically on
the quality and quantity of quantum-mechanical reference data with which they
are trained, and therefore developing datasets and training pipelines is
becoming an increasingly central challenge. Leveraging the idea of "synthetic"
(artificial) data that is common in other areas of ML research, we here show
that synthetic atomistic data, themselves obtained at scale with an existing ML
potential, constitute a useful pre-training task for neural-network interatomic
potential models. Once pre-trained with a large synthetic dataset, these models
can be fine-tuned on a much smaller, quantum-mechanical one, improving
numerical accuracy and stability in computational practice. We demonstrate
feasibility for a series of equivariant graph-neural-network potentials for
carbon, and we carry out initial experiments to test the limits of the
approach.
Related papers
- chemtrain: Learning Deep Potential Models via Automatic Differentiation and Statistical Physics [0.0]
Neural Networks (NNs) are promising models for refining the accuracy of molecular dynamics.
Chemtrain is a framework to learn sophisticated NN potential models through customizable training routines and advanced training algorithms.
arXiv Detail & Related papers (2024-08-28T15:14:58Z) - Discovering Interpretable Physical Models using Symbolic Regression and
Discrete Exterior Calculus [55.2480439325792]
We propose a framework that combines Symbolic Regression (SR) and Discrete Exterior Calculus (DEC) for the automated discovery of physical models.
DEC provides building blocks for the discrete analogue of field theories, which are beyond the state-of-the-art applications of SR to physical problems.
We prove the effectiveness of our methodology by re-discovering three models of Continuum Physics from synthetic experimental data.
arXiv Detail & Related papers (2023-10-10T13:23:05Z) - Capturing dynamical correlations using implicit neural representations [85.66456606776552]
We develop an artificial intelligence framework which combines a neural network trained to mimic simulated data from a model Hamiltonian with automatic differentiation to recover unknown parameters from experimental data.
In doing so, we illustrate the ability to build and train a differentiable model only once, which then can be applied in real-time to multi-dimensional scattering data.
arXiv Detail & Related papers (2023-04-08T07:55:36Z) - Transfer learning for chemically accurate interatomic neural network
potentials [0.0]
We show that pre-training the network parameters on data obtained from density functional calculations improves the sample efficiency of models trained on more accurate ab-initio data.
We provide GM-NN potentials pre-trained and fine-tuned on the ANI-1x and ANI-1ccx data sets, which can easily be fine-tuned on and applied to organic molecules.
arXiv Detail & Related papers (2022-12-07T19:21:01Z) - Synthetic data enable experiments in atomistic machine learning [0.0]
We demonstrate the use of a large dataset labelled with per-atom energies from an existing ML potential model.
The cheapness of this process, compared to the quantum-mechanical ground truth, allows us to generate millions of datapoints.
We show that learning synthetic data labels can be a useful pre-training task for subsequent fine-tuning on small datasets.
arXiv Detail & Related papers (2022-11-29T18:17:24Z) - Symbolic Regression in Materials Science: Discovering Interatomic
Potentials from Data [1.7149364927872015]
Machine learning can offset the high computational costs of ab initio atomic potentials.
symbolic regression is a powerful "white-box" approach for discovering functional forms of interatomic potentials.
Genetic programming-based approach for modeling atomic potentials is presented.
arXiv Detail & Related papers (2022-06-13T19:05:21Z) - EINNs: Epidemiologically-Informed Neural Networks [75.34199997857341]
We introduce a new class of physics-informed neural networks-EINN-crafted for epidemic forecasting.
We investigate how to leverage both the theoretical flexibility provided by mechanistic models as well as the data-driven expressability afforded by AI models.
arXiv Detail & Related papers (2022-02-21T18:59:03Z) - Equivariant vector field network for many-body system modeling [65.22203086172019]
Equivariant Vector Field Network (EVFN) is built on a novel equivariant basis and the associated scalarization and vectorization layers.
We evaluate our method on predicting trajectories of simulated Newton mechanics systems with both full and partially observed data.
arXiv Detail & Related papers (2021-10-26T14:26:25Z) - Quantum-tailored machine-learning characterization of a superconducting
qubit [50.591267188664666]
We develop an approach to characterize the dynamics of a quantum device and learn device parameters.
This approach outperforms physics-agnostic recurrent neural networks trained on numerically generated and experimental data.
This demonstration shows how leveraging domain knowledge improves the accuracy and efficiency of this characterization task.
arXiv Detail & Related papers (2021-06-24T15:58:57Z) - SE(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate
Interatomic Potentials [0.17590081165362778]
NequIP is a SE(3)-equivariant neural network approach for learning interatomic potentials from ab-initio calculations for molecular dynamics simulations.
The method achieves state-of-the-art accuracy on a challenging set of diverse molecules and materials while exhibiting remarkable data efficiency.
arXiv Detail & Related papers (2021-01-08T18:49:10Z) - Towards an Automatic Analysis of CHO-K1 Suspension Growth in
Microfluidic Single-cell Cultivation [63.94623495501023]
We propose a novel Machine Learning architecture, which allows us to infuse a neural deep network with human-powered abstraction on the level of data.
Specifically, we train a generative model simultaneously on natural and synthetic data, so that it learns a shared representation, from which a target variable, such as the cell count, can be reliably estimated.
arXiv Detail & Related papers (2020-10-20T08:36:51Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.