Multi-Dimensional Reconfigurable, Physically Composable Hybrid Diffractive Optical Neural Network

TL;DR

This paper introduces a multi-dimensional reconfigurable hybrid diffractive optical neural network (MDR-HDONN) that combines reconfigurability with ultra-parallel optical computing, enabling adaptable, high-speed, energy-efficient AI processing.

Contribution

The paper presents the first physically reconfigurable hybrid diffractive optical neural network architecture that leverages full-system learnability for enhanced versatility and task adaptability.

Findings

Achieves digital-comparable accuracy on various tasks

Operates 74x faster and uses 194x less energy than traditional methods

Offers 5x faster training speed and exponentially larger functional space

Abstract

Diffractive optical neural networks (DONNs), leveraging free-space light wave propagation for ultra-parallel, high-efficiency computing, have emerged as promising artificial intelligence (AI) accelerators. However, their inherent lack of reconfigurability due to fixed optical structures post-fabrication hinders practical deployment in the face of dynamic AI workloads and evolving applications. To overcome this challenge, we introduce, for the first time, a multi-dimensional reconfigurable hybrid diffractive ONN system (MDR-HDONN), a physically composable architecture that unlocks a new degree of freedom and unprecedented versatility in DONNs. By leveraging full-system learnability, MDR-HDONN repurposes fixed fabricated optical hardware, achieving exponentially expanded functionality and superior task adaptability through the differentiable learning of system variables. Furthermore, MDR-HDONN adopts a hybrid optical/photonic design, combining the reconfigurability of integrated photonics with the ultra-parallelism of free-space diffractive systems. Extensive evaluations demonstrate that MDR-HDONN has digital-comparable accuracy on various task adaptations with 74x faster speed and 194x lower energy. Compared to prior DONNs, MDR-HDONN shows exponentially larger functional space with 5x faster training speed, paving the way for a new paradigm of versatile, composable, hybrid optical/photonic AI computing. We will open-source our codes.