Online unsupervised Hebbian learning in deep photonic neuromorphic networks

TL;DR

This paper presents a novel all-optical deep photonic neuromorphic network that performs online, unsupervised Hebbian learning using phase-change material synapses, demonstrating high recognition accuracy on a letter recognition task.

Contribution

It introduces a purely photonic deep neural network architecture with optical feedback for Hebbian learning, enabling real-time, unsupervised learning without electrical conversions.

Findings

Achieved 100% letter recognition accuracy

Demonstrated online, unsupervised learning in a photonic system

Utilized phase-change materials for non-volatile synapses

Abstract

While software implementations of neural networks have driven significant advances in computation, the von Neumann architecture imposes fundamental limitations on speed and energy efficiency. Neuromorphic networks, with structures inspired by the brain's architecture, offer a compelling solution with the potential to approach the extreme energy efficiency of neurobiological systems. Photonic neuromorphic networks (PNNs) are particularly attractive because they leverage the inherent advantages of light, namely high parallelism, low latency, and exceptional energy efficiency. Previous PNN demonstrations have largely focused on device-level functionalities or system-level implementations reliant on supervised learning and inefficient optical-electrical-optical (OEO) conversions. Here, we introduce a purely photonic deep PNN architecture that enables online, unsupervised learning. We propose a local feedback mechanism operating entirely in the optical domain that implements a Hebbian learning rule using non-volatile phase-change material synapses. We experimentally demonstrate this approach on a non-trivial letter recognition task using a commercially available fiber-optic platform and achieve a 100 percent recognition rate, showcasing an all-optical solution for efficient, real-time information processing. This work unlocks the potential of photonic computing for complex artificial intelligence applications by enabling direct, high-throughput processing of optical information without intermediate OEO signal conversions.