Discovering Symmetry Invariants and Conserved Quantities by Interpreting Siamese Neural Networks

03/09/2020
by   Sebastian J. Wetzel, et al.
0

In this paper, we introduce interpretable Siamese Neural Networks (SNN) for similarity detectionto the field of theoretical physics. More precisely, we apply SNNs to events in special relativity, thetransformation of electromagnetic fields, and the motion of particles in a central potential. In theseexamples, these SNNs learn to identify datapoints belonging to the same events, field configurations, or trajectory of motion. It turns out that in the process of learning which datapoints belong to the same event or field configuration, these SNNs also learn the relevant symmetry invariants andconserved quantities. These SNNs are highly interpretable, which enables us to reveal the symmetry invariants and conserved quantities without prior knowledge.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/20/2017

Fast and Efficient Calculations of Structural Invariants of Chirality

Chirality plays an important role in physics, chemistry, biology, and ot...
research
04/16/2019

Learning a Gauge Symmetry with Neural Networks

We explore the capacity of neural networks to detect a symmetry with com...
research
01/24/2023

Autonomous particles

Consider a reinforcement learning problem where an agent has access to a...
research
06/10/2022

Finite electro-elasticity with physics-augmented neural networks

In the present work, a machine learning based constitutive model for ele...
research
04/29/2021

Improving Simulations with Symmetry Control Neural Networks

The dynamics of physical systems is often constrained to lower dimension...
research
01/12/2022

Reduced polynomial invariant integrity basis for in-plane magneto-mechanical loading

The description of the behavior of a material subjected to multi-physics...
research
07/13/2021

Spontaneous Symmetry Breaking for Extreme Vorticity and Strain in the 3D Navier-Stokes Equations

We investigate the spatio-temporal structure of the most likely configur...

Please sign up or login with your details

Forgot password? Click here to reset