# Machine-Learning Non-Conservative Dynamics for New-Physics Detection

@article{Liu2021MachineLearningND, title={Machine-Learning Non-Conservative Dynamics for New-Physics Detection}, author={Ziming Liu and Bohan Wang and Qi Meng and Wei Chen and Max Tegmark and Tie-Yan Liu}, journal={ArXiv}, year={2021}, volume={abs/2106.00026} }

Energy conservation is a basic physics principle, the breakdown of which often implies new physics. This paper presents a method for data-driven “new physics” discovery. Specifically, given a trajectory governed by unknown forces, our Neural New-Physics Detector (NNPhD) aims to detect new physics by decomposing the force field into conservative and non-conservative components, which are represented by a Lagrangian Neural Network (LNN) and a universal approximator network (UAN), respectively… Expand

#### Figures and Tables from this paper

#### 2 Citations

Extracting Dynamical Models from Data

- Computer Science
- ArXiv
- 2021

The FJet approach is introduced for determining the underlying model of a dynamical system, which naturally overcomes the "extrapolation problem", by doing the modeling in the phase space of the system, rather than over the time domain. Expand

Equivariant vector field network for many-body system modeling

- Computer Science, Mathematics
- ArXiv
- 2021

The Equivariant Vector Field Network (EVFN), which is built on a novel tuple of equivariant basis and the associated scalarization and vectorization layers, is proposed, which achieves best or competitive performance on baseline models in various types of datasets. Expand

#### References

SHOWING 1-10 OF 30 REFERENCES

Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

- Computer Science
- J. Comput. Phys.
- 2019

Abstract We introduce physics-informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given laws of physics described by general nonlinear… Expand

Toward an artificial intelligence physicist for unsupervised learning.

- Medicine, Physics
- Physical review. E
- 2019

This work proposes a paradigm centered around the learning and manipulation of theories, which parsimoniously predict both aspects of the future and the domain in which these predictions are accurate, and proposes a generalized mean loss to encourage each theory to specialize in its comparatively advantageous domain. Expand

Galileo: Perceiving Physical Object Properties by Integrating a Physics Engine with Deep Learning

- Computer Science
- NIPS
- 2015

This study points towards an account of human vision with generative physical knowledge at its core, and various recognition models as helpers leading to efficient inference. Expand

The basic physics of the binary black hole merger GW150914

- Physics
- 2017

The first direct gravitational-wave detection was made by the Advanced Laser Interferometer Gravitational Wave Observatory on September 14, 2015. The GW150914 signal was strong enough to be apparent,… Expand

Forecasting Hamiltonian dynamics without canonical coordinates

- Computer Science, Physics
- ArXiv
- 2020

This work prepend a conventional neural network to a Hamiltonian neural network and shows that the combination accurately forecasts Hamiltonian dynamics from generalised noncanonical coordinates. Expand

Simplifying Hamiltonian and Lagrangian Neural Networks via Explicit Constraints

- Computer Science, Mathematics
- NeurIPS
- 2020

This paper introduces a series of challenging chaotic and extended-body systems, including systems with N-pendulums, spring coupling, magnetic fields, rigid rotors, and gyroscopes, and shows that embedding the system into Cartesian coordinates and enforcing the constraints explicitly with Lagrange multipliers dramatically simplifies the learning problem. Expand

Extracting Interpretable Physical Parameters from Spatiotemporal Systems using Unsupervised Learning

- Computer Science, Physics
- ArXiv
- 2019

This work implements a physics-informed architecture based on variational autoencoders that is designed for analyzing systems governed by partial differential equations (PDEs) and extracts latent parameters that parameterize the dynamics of a learned predictive model for the system. Expand

Why Does Deep and Cheap Learning Work So Well?

- Mathematics, Physics
- ArXiv
- 2016

It is argued that when the statistical process generating the data is of a certain hierarchical form prevalent in physics and machine learning, a deep neural network can be more efficient than a shallow one. Expand

AI Poincaré: Machine Learning Conservation Laws from Trajectories

- Medicine, Computer Science
- Physical review letters
- 2021

AI Poincaré, a machine learning algorithm for autodiscovering conserved quantities using trajectory data from unknown dynamical systems, is presented and it is found that it discovers not only all exactly conserved quantity, but also periodic orbits, phase transitions, and breakdown timescales for approximate conservation laws. Expand

Hamiltonian Neural Networks

- Computer Science
- NeurIPS
- 2019

Inspiration from Hamiltonian mechanics is drawn to train models that learn and respect exact conservation laws in an unsupervised manner, and this model trains faster and generalizes better than a regular neural network. Expand