Multiple delay nonlinear wavelength dynamics for photonic Reservoir Computing
Autor: | Maxime Jacquot, Sergei Rybalko, Romain Martinenghi, Laurent Larger, Lennert Appeltant, G. Van der Sand, Jan Danckaert, Yanne K. Chembo |
---|---|
Přispěvatelé: | Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) (FEMTO-ST), Université de Technologie de Belfort-Montbeliard (UTBM)-Ecole Nationale Supérieure de Mécanique et des Microtechniques (ENSMM)-Université de Franche-Comté (UFC), Université Bourgogne Franche-Comté [COMUE] (UBFC)-Université Bourgogne Franche-Comté [COMUE] (UBFC)-Centre National de la Recherche Scientifique (CNRS), Applied Physics Research Group - Vrije Universiteit Brussel (APHY), Vrije Universiteit Brussel (VUB) |
Rok vydání: | 2011 |
Předmět: |
Artificial neural network
Computer science business.industry Frame (networking) Reservoir computing Optical computing 01 natural sciences Field (computer science) 03 medical and health sciences 0302 clinical medicine 0103 physical sciences [SPI.OPTI]Engineering Sciences [physics]/Optics / Photonic Electronic engineering Photonics 010306 general physics business Implementation Realization (systems) 030217 neurology & neurosurgery |
Zdroj: | Lasers and Electro-Optics Europe (CLEO EUROPE/EQEC) 2011 Conference on and 12th European Quantum Electronics 2011 Conference on and 12th European Quantum Electronics, May 2011, Munich, Germany. pp.EH4_5, ⟨10.1109/CLEOE.2011.5943557⟩ Scopus-Elsevier |
Popis: | International audience; Reservoir computing is an emerging field of research intended to develop novel computational machines based on operating principles issued from the neural network community and the brain research sciences. Practical implementations of the corresponding principles is a strongly open question, which is precisely addressed from the photonic viewpoint in the frame of a recently funded EC project PHOCUS. The main idea is here to benefit from the intrinsic dynamic complexity of delay systems, which experimental realization are well known in photonics, in order to perform the central role of a reservoir computer, the reservoir itself. This paper reports on the photonic implementation, the numerical simulations of the corresponding reservoir subjected to the NARMA sequences, and the computational power results that were obtained both numerically and experimentally. |
Databáze: | OpenAIRE |
Externí odkaz: |