Parallel Extreme Learning Machines Based on Frequency Multiplexing
Source abstract
Abstract
In a recent work, we reported on an Extreme Learning Machine (ELM) implemented in a photonic system based on frequency multiplexing, where each wavelength of the light encodes a different neuron state. In the present work, we experimentally demonstrate the parallelization potentialities of this approach. We show that multiple frequency combs centered on different frequencies can copropagate in the same system, resulting in either multiple independent ELMs executed in parallel on the same substrate or a single ELM with an increased number of neurons. We experimentally tested the performances of both these operation modes on several classification tasks, employing up to three different light sources, each of which generates an independent frequency comb. We also numerically evaluated the performances of the system in configurations containing up to 15 different light sources.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Neural Networks and Reservoir Computing · Advanced Memory and Neural Computing · Machine Learning and ELM
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Laboratoire d’Information Quantique, CP 224, Université Libre de Bruxelles, Av. F. D. Roosevelt 50, B-1050 Bruxelles, Belgium
Funding
European Commission · Fonds de la Recherche Scientifique (FRS-FNRS)
References (32)
Show source-provided references
1. Randomness in Neural Networks: An Overview10.1002/widm.1200 ↗
2. Huang, G.B., Zhu, Q.Y., and Siew, C.K. (2004, January 25–29). Extreme learning machine: A new learning scheme of feedforward neural networks. Proceedings of the IEEE International Joint Conference on Neural Networks, Budapest, Hungary.
3. Extreme learning machine: Theory and applications10.1016/j.neucom.2005.12.126 ↗
4. What are extreme learning machines? Filling the gap between Frank Rosenblatt’s dream and John von Neumann’s puzzle10.1007/s12559-015-9333-0 ↗
5. Gonon, L., Grigoryeva, L., and Ortega, J.P. (2002). Approximation bounds for random neural networks and reservoir systems. arXiv.
6. A survey of approaches for implementing optical neural networks10.1016/j.optlastec.2020.106787 ↗
7. Saade, A., Caltagirone, F., Carron, I., Daudet, L., Drémeau, A., Gigan, S., and Krzakala, F. (2016, January 20–25). Random projections through multiple optical scattering: Approximating Kernels at the speed of light. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, Shanghai, China.10.1109/ICASSP.2016.7472872 ↗
8. Photonic extreme learning machine by free-space optical propagation10.1364/PRJ.423531 ↗
9. Tegin, U., Yıldırım, M., Oguz, I., Moser, C., and Psaltis, D. (2012). Scalable Optical Learning Operator. arXiv.
10. A unified framework for reservoir computing and extreme learning machines based on a single time-delayed neuron10.1038/srep14945 ↗
11. Information processing using a single dynamical node as complex system10.1038/ncomms1476 ↗
12. Optoelectronic reservoir computing10.1038/srep00287 ↗
13. Photonic information processing beyond Turing: An optoelectronic implementation of reservoir computing10.1364/OE.20.003241 ↗
14. Fyfe, C., Tino, P., Charles, D., Garcia-Osorio, C., and Yin, H. (2010). Simple Deterministically Constructed Recurrent Neural Networks. Intelligent Data Engineering and Automated Learning—IDEAL 2010, Springer.10.1007/978-3-642-15381-5 ↗
15. Reinforcement learning in a large-scale photonic recurrent neural network10.1364/OPTICA.5.000756 ↗
16. Human action recognition with a large-scale brain-inspired photonic computer10.1038/s42256-019-0110-8 ↗
17. Experimental demonstration of reservoir computing on a silicon photonics chip10.1038/ncomms4541 ↗
18. Butschek, L., Akrout, A., Dimitriadou, E., Haelterman, M., and Massar, S. (2008). Parallel photonic reservoir computing based on frequency multiplexing of neurons. arXiv.
19. Photonic extreme learning machine based on frequency multiplexing10.1364/OE.433535 ↗
20. Photonic Perceptron Based on a Kerr Microcomb for High-Speed, Scalable, Optical Neural Networks10.1002/lpor.202000070 ↗
21. 11 TOPS photonic convolutional accelerator for optical neural networks10.1038/s41586-020-03063-0 ↗
22. Parallel convolutional processing using an integrated photonic tensor core10.1038/s41586-020-03070-1 ↗
23. A Parallel Multiclassification Algorithm for Big Data Using an Extreme Learning Machine10.1109/TNNLS.2017.2654357 ↗
24. Parallel photonic information processing at gigabyte per second data rates using transient states10.1038/ncomms2368 ↗
25. Virtualization of a Photonic Reservoir Computer10.1109/JLT.2016.2524559 ↗
26. Electro-optic frequency combs10.1364/AOP.382052 ↗
27. Fisher, R. (2021, November 15). Iris Data Set. Available online: http://archive.ics.uci.edu/ml/datasets/iris.
28. Forina, M. (2021, November 15). Wine Data Set. Available online: https://archive.ics.uci.edu/ml/datasets/wine.
29. Lohweg, V. (2021, November 15). Banknote Authentication Data Set. Available online: https://archive.ics.uci.edu/ml/datasets/banknote+authentication.
30. Lincoff, G.H. (2021, November 15). Mushroom Data Set. Available online: https://archive.ics.uci.edu/ml/datasets/mushroom.
31. LeCun, Y., Cortes, C., and Burges, C.J.C. (2021, November 15). MNIST Database of Handwritten Digits. Available online: http://yann.lecun.com/exdb/mnist/.
32. de Chazal, P., Tapson, J., and van Schaik, A. (2015, January 19–24). A comparison of extreme learning machines and back-propagation trained feed-forward networks processing the mnist database. Proceedings of the 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), South Brisbane, Australia.10.1109/ICASSP.2015.7178354 ↗