Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems
This work addresses the energy and scaling limitations of deep neural networks for broader applications in science and engineering, offering a radical alternative that could enable more efficient hardware.
The authors tackled the growing energy inefficiency of deep neural networks by proposing Physical Neural Networks, which use controllable physical systems as network layers, and demonstrated their approach with optical, mechanical, and electrical systems, achieving potential orders-of-magnitude improvements in speed and energy efficiency.
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.