Optical Implementation of Equilibrium Propagation Using Spatial Photonic Ising Machines
For researchers in neuromorphic computing and optical neural networks, this work offers a proof-of-concept for physically implementing Equilibrium Propagation, though it is incremental as it only demonstrates on a small dataset and relies on numerical simulations for larger tasks.
The authors demonstrate a hybrid optical-digital implementation of Equilibrium Propagation using a Spatial Photonic Ising Machine, achieving classification on the Wine dataset experimentally and evaluating potential on MNIST numerically, providing a pathway toward energy-efficient physical neural networks.
Equilibrium Propagation offers a compelling alternative to traditional machine learning for training energy-based networks. Here we demonstrate a hybrid optical-digital implementation of EP using a Spatial Photonic Ising Machine (SPIM). The SPIM exploits the gauge transformation method to optically encode both continuous neuron states and rank-1 binary trainable patterns as phase modulations via a spatial light modulator, with inference realized using a finite difference scheme. The experimental system is evaluated on the Wine classification dataset. The potential of this approach, including the use of continuous couplings and structured coupling matrices, is evaluated numerically on the more complex MNIST dataset. Our work provides a concrete pathway toward energy-efficient physical implementations of Equilibrium Propagation.