Photonic reservoir computer based on frequency multiplexing

Abstract

Reservoir computing is a brain inspired approach for information processing, well suited to analogue implementations. We report a photonic implementation of a reservoir computer that exploits frequency domain multiplexing to encode neuron states. The system processes 25 comb lines simultaneously (i.e. 25 neurons), at a rate of 20 MHz. We illustrate performances on two standard benchmark tasks: channel equalization and time series forecasting. We also demonstrate that frequency multiplexing allows output weights to be implemented in the optical domain, through optical attenuation. We discuss the perspectives for high speed high performance low footprint implementations.

19 pages. Published version, including supplementary material

References in corpus (4)

• 11 TeraFLOPs per second photonic convolutional accelerator for deep learning optical neural networks
• High performance photonic reservoir computer based on a coherently driven passive cavity
• Classification of time-domain waveforms using a speckle-based optical reservoir computer
• Photonic Extreme Learning Machine based on frequency multiplexing

Cited by in corpus (7)

• Pattern recognition with a magnon-scattering reservoir
• Deep Photonic Reservoir Computer Based on Frequency Multiplexing with Fully Analog Connection Between Layers
• Squeezing as a resource for time series processing in quantum reservoir computing
• Optical neuromorphic computing via temporal up-sampling and trainable encoding on a telecom device platform
• Thermally-robust spatiotemporal parallel reservoir computing by frequency filtering in frustrated magnets
• Optimizing the Network Topology of a Linear Reservoir Computer
• Shannon Entropy Helps Optimize the Performance of a Frequency-Multiplexed Extreme Learning Machine