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

Abstract

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, trained to minimize the force recovery error plus a constant $\lambda$ times the magnitude of the predicted non-conservative force. We show that a phase transition occurs at $\lambda$=1, universally for arbitrary forces. We demonstrate that NNPhD successfully discovers new physics in toy numerical experiments, rediscovering friction (1493) from a damped double pendulum, Neptune from Uranus' orbit (1846) and gravitational waves (2017) from an inspiraling orbit. We also show how NNPhD coupled with an integrator outperforms previous methods for predicting the future of a damped double pendulum.

arXiv comment17 pages, 7 figs, 2 tables; typo correction

References in corpus (14)

• Data-driven discovery of coordinates and governing equations
• Hamiltonian Neural Networks
• Why does deep and cheap learning work so well?
• Discovering Symbolic Models from Deep Learning with Inductive Biases
• Integration of Neural Network-Based Symbolic Regression in Deep Learning for Scientific Discovery
• AI Poincaré: Machine Learning Conservation Laws from Trajectories
• The basic physics of the binary black hole merger GW150914
• Toward an AI Physicist for Unsupervised Learning
• Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning
• Lagrangian Neural Networks
• Symbolic Pregression: Discovering Physical Laws from Distorted Video
• Dissipative SymODEN: Encoding Hamiltonian Dynamics with Dissipation and Control into Deep Learning
• Extracting Interpretable Physical Parameters from Spatiotemporal Systems using Unsupervised Learning
• Neural Canonical Transformation with Symplectic Flows

Cited by in corpus (3)

• Machine-learning hidden symmetries
• Learning phase transitions from regression uncertainty: A new regression-based machine learning approach for automated detection of phases of matter
• Extracting Dynamical Models from Data