Comprehensive Performance Analysis of a VCSEL-Based Photonic Reservoir Computer

Bueno, Julián; Robertson, Joshua; Hejda, Matěj; Hurtado, Antonio

doi:10.1109/lpt.2021.3075095

Cited by 34 publications

(18 citation statements)

References 19 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3(c,d). The authors find that ROT feedback degrades the computational performance in terms of memory capacity when there is a significant power difference between the two emission modes 27 .…”

Section: Time Delayed Reservoirsmentioning

confidence: 99%

“…Vatin et al numerically 30 and experimentally 15 first demonstrated delaybased RC with two-mode polarization dynamics of a VC-SEL, which offers a bigger playground for optical feedback and injection configurations. Using either parallel (PAR) or orthogonal (ROT) configurations, a comprehensive experimental analysis can be found in 27 , where the authors analyze the performance of both feedback configurations with Mackey-Glass prediction and nonlinear channel equalization tasks, see Fig. 3(c,d).…”

Section: Time Delayed Reservoirsmentioning

confidence: 99%

See 1 more Smart Citation

Photonic neuromorphic computing using vertical cavity semiconductor lasers

Skalli¹,

Robertson²,

Owen-Newns³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Section: Time Delayed Reservoirsmentioning

confidence: 99%

Section: Time Delayed Reservoirsmentioning

confidence: 99%

Photonic neuromorphic computing using vertical cavity semiconductor lasers

Skalli¹,

Robertson²,

Owen-Newns³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

“…In this approach, the input signals entering the VCSEL-based SNN (spiking photonic RC system) are chosen (for this proof-of-concept demonstration) to be analogous to those used previously in traditional non-spiking laser-based photonic RC systems (i.e. continuously modulated light input) [14]. Yet, we believe operation with other types of input signals (such as pulsed or spiking inputs) should also be possible and will be investigated in the future.…”

Section: A Vcsel-based Snn Structure and Experimental Setupmentioning

confidence: 99%

“…ELMs are purely feed-forward neural networks that operate on the same principles of a fixed reservoir of nodes, and a single output layer that needs training. The simplified architecture and training procedure makes the implementation of RC/ELM systems in photonic hardware highly desirable, and now photonic platforms based on numerous devices [20], [21], [22], [14], [15] (also recently including VCSELs), have successfully performed information processing tasks such as image classification, time series prediction and signal conditioning with state of the art performance. However, to date neuromorphic photonic systems, such as laser-based photonic reservoir computers, have not demonstrated spiking neural network (SNN) architectures using excitable neural-like spikes for information processing.…”

Section: Introductionmentioning

confidence: 99%

GHz Rate Neuromorphic Photonic Spiking Neural Network With a Single Vertical-Cavity Surface-Emitting Laser (VCSEL)

Owen-Newns

Robertson

Hejda

et al. 2023

IEEE J. Select. Topics Quantum Electron.

Self Cite

View full text Add to dashboard Cite

Vertical-Cavity Surface-Emitting Lasers (VCSELs) are highly promising devices for the construction of neuromorphic photonic information processing systems, due to their numerous desirable properties such as low power consumption, high modulation speed, compactness, and ease of manufacturing. Of particular interest is the ability of VCSELs to exhibit neurallike spiking responses, much like biological neurons, but at ultrafast sub-nanosecond rates; thus offering great prospects for high-speed light-enabled neuromorphic (spike-based) processors. Recent works have shown the use the spiking dynamics in VC-SELs for pattern recognition and image processing problems such as image data encoding and edge-feature detection. Additionally, VCSELs have also been used recently as nonlinear elements in photonic reservoir computing (RC) implementations, yielding excellent state of the art operation. This work introduces and experimentally demonstrates for the first time the new concept of a Ghz-rate photonic spiking neural network (SNN) built with a single VCSEL neuron. The reported system effectively implements a photonic VCSEL-based spiking reservoir computer, and demonstrates its successful application to a complex nonlinear classification task. Importantly, the proposed system benefits from a highly hardware-friendly, inexpensive realization (built with a single VCSEL and off-the-shelf fibre-optic components), for high-speed (GHz-rate inputs) and low-power (sub-mW optical input power) photonic operation. These results open new pathways towards future neuromorphic photonic spikebased information processing systems based upon VCSELs (or other laser types) for novel ultrafast machine learning and AI hardware.

show abstract

“…Among diverse reservoir computing implementation methods, photonic reservoir computing is one of the most prospective methods due to the fast information processing speed, multiple multiplexing capabilities and ultra-low power consumption [23][24][25][26]. At present, vast amounts of photonic reservoir computing are concentrated on time-delayed RC architecture with semiconductor lasers (SLs) [20,23,[27][28][29][30], such as quantum dot laser [23], Fabry-Perot laser [20], and two-element phased laser array [27]. Intriguingly, with the progress of nanophotonic technology, it is particularly compelling that nanophotonic devices have incomparable advantages over semiconductor lasers in terms of scalability, com-plementary metal oxide semiconductor (CMOS) compatibility, monolithic integration and mass production [22,[31][32][33].…”

Section: Introductionmentioning

confidence: 99%