Virtual reservoir computer using an optical resonator
Source abstract
Abstract
Reservoir computing is a machine learning approach that enables us to use recurrent neural networks without involving the complexity of training algorithms and make hardware implementation possible. We present a novel photonic architecture of a reservoir computer that employs a nonlinear node and a resonator to implement a virtual recurrent neural network. This resonator behaves as an echo generator component that substitutes the delay line in delay-based reservoir computers available in the literature. The virtual neural network formed in our implementation is fundamentally different from the delay-based reservoir computers. Different virtual architectures based on the FSR and the Finesse of the resonator are investigated to provide higher performance depending on the task. We test the performance of our proposed reservoir by 3 benchmark tasks, signal classification, nonlinear channel equalization, and memory capacity. Our system enhances the memory capacity of the reservoir compared to the delay-based systems and shows reliable results in signal classification and nonlinear channel equalization.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Neural Networks and Reservoir Computing · Optical Network Technologies · Advanced Memory and Neural Computing
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Sharif University of Technology
References (24)
Show source-provided references
1. Maass · Neural Comput. · 2002 · 14 · 253110.1162/089976602760407955 ↗
2. Caulfield · Nat. Photonics · 2010 · 4 · 26110.1038/nphoton.2010.94 ↗
3. Tucker · Nat. Photonics · 2010 · 4 · 40510.1038/nphoton.2010.162 ↗
4. Miller · Nat. Photonics · 2010 · 4 · 40610.1038/nphoton.2010.163 ↗
5. Woods · Nat. Phys. · 2012 · 8 · 25710.1038/nphys2283 ↗
6. Vandoorne · Opt. Express · 2008 · 16 · 1118210.1364/OE.16.011182 ↗
7. Vandoorne · IEEE Trans. Neural Netw. · 2011 · 22 · 146910.1109/TNN.2011.2161771 ↗
8. Denis-Le Coarer · IEEE J. Sel. Top. Quantum Electron. · 2018 · 24 · 110.1109/JSTQE.2018.2836985 ↗
9. Vandoorne · Nat. Commun. · 2014 · 5 · 354110.1038/ncomms4541 ↗
10. Fiers · IEEE Trans. Neural Netw. Learning Syst. · 2014 · 25 · 34410.1109/TNNLS.2013.2274670 ↗
11. Paquot · Sci. Rep. · 2012 · 2 · 28710.1038/srep00287 ↗
12. Reservoir computing: a photonic neural network for information processing
13. Appeltant · Nat. Commun. · 2011 · 2 · 46810.1038/ncomms1476 ↗
14. Larger · Opt. Express · 2012 · 20 · 324110.1364/OE.20.003241 ↗
15. Duport · Opt. Express · 2012 · 20 · 2278310.1364/OE.20.022783 ↗
16. Dejonckheere · Opt. Express · 2014 · 22 · 1086810.1364/OE.22.010868 ↗
17. Brunner · Nat. Commun. · 2013 · 4 · 136410.1038/ncomms2368 ↗
18. Hicke · IEEE J. Sel. Top. Quantum Electron. · 2013 · 19 · 150161010.1109/JSTQE.2013.2241738 ↗
19. Nguimdo · Opt. Express · 2014 · 22 · 867210.1364/OE.22.008672 ↗
20. Nguimdo · IEEE Trans. Neural Netw. Learning Syst. · 2015 · 26 · 330110.1109/TNNLS.2015.2404346 ↗
21. Vinckier · Optica · 2015 · 2 · 43810.1364/OPTICA.2.000438 ↗
22. Zhang · Opt. Express · 2014 · 22 · 3135610.1364/OE.22.031356 ↗
23. Takano · Opt. Express · 2018 · 26 · 2942410.1364/OE.26.029424 ↗
24. Harkhoe · Opt. Express · 2020 · 28 · 308610.1364/OE.382556 ↗