Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems

04/27/2021
by   Logan G. Wright, et al.
77

Deep neural networks have become a pervasive tool in science and engineering. However, modern deep neural networks' growing energy requirements now increasingly limit their scaling and broader use. We propose a radical alternative for implementing deep neural network models: Physical Neural Networks. We introduce a hybrid physical-digital algorithm called Physics-Aware Training to efficiently train sequences of controllable physical systems to act as deep neural networks. This method automatically trains the functionality of any sequence of real physical systems, directly, using backpropagation, the same technique used for modern deep neural networks. To illustrate their generality, we demonstrate physical neural networks with three diverse physical systems-optical, mechanical, and electrical. Physical neural networks may facilitate unconventional machine learning hardware that is orders of magnitude faster and more energy efficient than conventional electronic processors.

READ FULL TEXT

page 4

page 7

page 26

page 30

page 35

page 38

page 40

page 41

research
04/20/2023

Backpropagation-free Training of Deep Physical Neural Networks

Recent years have witnessed the outstanding success of deep learning in ...
research
01/02/2020

Operationally meaningful representations of physical systems in neural networks

To make progress in science, we often build abstract representations of ...
research
03/18/2021

A deep learning theory for neural networks grounded in physics

In the last decade, deep learning has become a major component of artifi...
research
11/15/2018

Physical Signal Classification Via Deep Neural Networks

A Deep Neural Network is applied to classify physical signatures obtaine...
research
06/23/2020

Learning Physical Constraints with Neural Projections

We propose a new family of neural networks to predict the behaviors of p...
research
02/27/2021

Characterization of Neural Networks Automatically Mapped on Automotive-grade Microcontrollers

Nowadays, Neural Networks represent a major expectation for the realizat...
research
01/26/2022

Voronoi cell analysis: The shapes of particle systems

Many physical systems can be studied as collections of particles embedde...

Please sign up or login with your details

Forgot password? Click here to reset