Visualizing a neural network that develops quantum perturbation theory

Motivated by the question whether the empirical fitting of data by neural networks can yield the same structure of physical laws, we apply neural networks to a quantum-mechanical two-body scattering problem with short-range potentials—a problem that by itself plays an important role in many branches...

Full description

Bibliographic Details
Main Authors: Wu, Yadong, Zhang, Pengfei, Shen, Huitao, Zhai, Hui
Other Authors: Massachusetts Institute of Technology. Department of Physics
Format: Article
Language:English
Published: American Physical Society 2018
Online Access:http://hdl.handle.net/1721.1/117207
https://orcid.org/0000-0003-1667-8011
_version_ 1811078486889070592
author Wu, Yadong
Zhang, Pengfei
Shen, Huitao
Zhai, Hui
author2 Massachusetts Institute of Technology. Department of Physics
author_facet Massachusetts Institute of Technology. Department of Physics
Wu, Yadong
Zhang, Pengfei
Shen, Huitao
Zhai, Hui
author_sort Wu, Yadong
collection MIT
description Motivated by the question whether the empirical fitting of data by neural networks can yield the same structure of physical laws, we apply neural networks to a quantum-mechanical two-body scattering problem with short-range potentials—a problem that by itself plays an important role in many branches of physics. After training, the neural network can accurately predict s-wave scattering length, which governs the low-energy scattering physics. By visualizing the neural network, we show that it develops perturbation theory order by order when the potential depth increases, without solving the Schrödinger equation or obtaining the wave function explicitly. The result provides an important benchmark to the machine-assisted physics research or even automated machine learning physics laws.
first_indexed 2024-09-23T11:00:44Z
format Article
id mit-1721.1/117207
institution Massachusetts Institute of Technology
language English
last_indexed 2024-09-23T11:00:44Z
publishDate 2018
publisher American Physical Society
record_format dspace
spelling mit-1721.1/1172072022-10-01T00:34:22Z Visualizing a neural network that develops quantum perturbation theory Wu, Yadong Zhang, Pengfei Shen, Huitao Zhai, Hui Massachusetts Institute of Technology. Department of Physics Shen, Huitao Motivated by the question whether the empirical fitting of data by neural networks can yield the same structure of physical laws, we apply neural networks to a quantum-mechanical two-body scattering problem with short-range potentials—a problem that by itself plays an important role in many branches of physics. After training, the neural network can accurately predict s-wave scattering length, which governs the low-energy scattering physics. By visualizing the neural network, we show that it develops perturbation theory order by order when the potential depth increases, without solving the Schrödinger equation or obtaining the wave function explicitly. The result provides an important benchmark to the machine-assisted physics research or even automated machine learning physics laws. 2018-07-31T12:05:34Z 2018-07-31T12:05:34Z 2018-07 2018-03 2018-07-30T16:18:11Z Article http://purl.org/eprint/type/JournalArticle 2469-9926 2469-9934 http://hdl.handle.net/1721.1/117207 Wu, Yadong, Pengfei Zhang, Huitao Shen and Hui Zhai. "Visualizing a neural network that develops quantum perturbation theory." Physucial Review A 98 (2018), 010701. https://orcid.org/0000-0003-1667-8011 en http://dx.doi.org/10.1103/PhysRevA.98.010701 Physical Review A Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. American Physical Society application/pdf American Physical Society American Physical Society
spellingShingle Wu, Yadong
Zhang, Pengfei
Shen, Huitao
Zhai, Hui
Visualizing a neural network that develops quantum perturbation theory
title Visualizing a neural network that develops quantum perturbation theory
title_full Visualizing a neural network that develops quantum perturbation theory
title_fullStr Visualizing a neural network that develops quantum perturbation theory
title_full_unstemmed Visualizing a neural network that develops quantum perturbation theory
title_short Visualizing a neural network that develops quantum perturbation theory
title_sort visualizing a neural network that develops quantum perturbation theory
url http://hdl.handle.net/1721.1/117207
https://orcid.org/0000-0003-1667-8011
work_keys_str_mv AT wuyadong visualizinganeuralnetworkthatdevelopsquantumperturbationtheory
AT zhangpengfei visualizinganeuralnetworkthatdevelopsquantumperturbationtheory
AT shenhuitao visualizinganeuralnetworkthatdevelopsquantumperturbationtheory
AT zhaihui visualizinganeuralnetworkthatdevelopsquantumperturbationtheory