Spiking VCSEL-neuron. (a) Illustration of a biological neuron. (b) Experimental setup used to investigate a spiking VCSEL-neuron under external optical injection of intensity-encoded stimuli. Optical fibre connections are shown in red and electrical connections in blue. (c) Idealistic depiction of the LIF model. Inputs injected into the device (In) are integrated (Int.), with a time constant decay, towards a threshold potential (red dotted line). When the threshold requirement is met, the system fires a spiking response (Out) and the potential reaches the reset value (dark red) before returning to its resting potential (light blue). (d) Flow diagram of the VCSEL-neuron. Optical injection is encoded with pre-weighted inputs. These are integrated over time in the VCSEL-neuron where a spike activation function thresholds inputs before firing.
In today’s data-driven world, the ability to process large data volumes is crucial. Key tasks, such as pattern recognition and image classification, are well suited for artificial neural networks (ANNs) inspired by the brain. Neuromorphic computing approaches aimed towards physical realizations of ANNs
have been traditionally supported by micro-electronic platforms, but recently, photonic techniques for neuronal emulation have emerged given their unique properties (e.g. ultrafast operation, large bandwidths, low cross-talk). Yet, hardware-friendly systems of photonic spiking neurons able to perform processing tasks at high speeds and with continuous operation remain elusive. This work provides a first experimental report of Vertical-Cavity Surface-Emitting Laser-based spiking neurons demonstrating different functional processing tasks, including coincidence detection and pattern recognition, at ultrafast rates. Furthermore, our approach relies on simple hardware implementations using off-the-shelf components. These results therefore hold exciting prospects for novel, compact and high-speed neuromorphic photonic platforms for future computing and Artificial Intelligence systems.
The ChipAI partner University of Strathclyde that published the paper “Ultrafast optical integration and pattern classification for neuromorphic photonics based on spiking VCSEL neurons” in Scientific Reports (Nature Publishing). The paper reports pattern recognition at sub-ns bit rates using a photonic spiking VCSEL-Neuron. These results hold exciting prospects of functional processing tasks in the context of high-speed neuromorphic photonic platforms for future computing and artificial intelligence systems.
