This paper investigates how network architecture, specifically locally connected lattices, impacts the success of Equilibrium Propagation (EP) training in neuromorphic systems. It demonstrates that sparse networks with local connections can achieve performance comparable to dense networks, offering guidelines for scaling up EP-based architectures in realistic settings.

This paper introduces an optical spiking neural network that utilizes optical rogue-wave statistics as a programmable firing mechanism. It demonstrates how phase-engineered caustics enable robust, passive thresholding, thereby harnessing extreme-wave phenomena for scalable and energy-efficient neuromorphic photonic inference.