Authors:	K.T Vandoorne, J. Dambre, D. Verstraeten, B. Schrauwen, P. Bienstman
Title:	Delays in photonic reservoir computing with Semiconductor Optical Amplifiers
Format:	International Conference Proceedings
Publication date:	11/2010
Journal/Conference/Book:	IEEE Photonics Benelux Chapter Annual Symposium
Volume(Issue):	p.25-28
Location:	Delft, Netherlands
Citations:	Look up on Google Scholar
Download:	(226KB)

Abstract

Reservoir Computing is a methodology coming from the field of machine learning and
neural networks that has been successfully used in pattern classification problems. We
have investigated with simulations a network of coupled semiconductor optical amplifiers
as an integrated photonic reservoir on a benchmark speech recognition task. In spite of
the differences with classical software reservoir models, the performance of our photonic
reservoir is comparable to that of conventional implementations with the same number of
nodes. It performs even better when the most important parameter, the amount of delay
in the interconnections, is tuned to its optimal value.

Related Research Topics

• Photonic Reservoir Computing

Related Projects

• ERC-StG: NaResCo (Novel paradigms for massively parallel nanophotonic information processing: photonics reservoir computing)