Bayesian Optimisation of Large-scale Photonic Reservoir Computers - Université de Lorraine Accéder directement au contenu
Article Dans Une Revue Cognitive Computation Année : 2021

Bayesian Optimisation of Large-scale Photonic Reservoir Computers

Résumé

Reservoir computing is a growing paradigm for simplified training of recurrent neural networks, with a high potential for hardware implementations. Numerous experiments in optics and electronics yield comparable performance with digital state-of-the-art algorithms. Many of the most recent works in the field focus on large-scale photonic systems, with tens of thousands of physical nodes and arbitrary interconnections. While this trend significantly expands the potential applications of photonic reservoir computing, it also complicates the optimisation of the high number of hyper-parameters of the system. In this work, we propose the use of Bayesian optimisation for efficient exploration of the hyper-parameter space in a minimum number of iteration. We test this approach on a previously reported large-scale experimental system, compare it with the commonly used grid search, and report notable improvements in performance and the number of experimental iterations required to optimise the hyper-parameters. Bayesian optimisation thus has the potential to become the standard method for tuning the hyper-parameters in photonic reservoir computing.
Fichier principal
Vignette du fichier
be20b913-35dc-4753-98b7-d5518bead729-author.pdf (582.48 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03360041 , version 1 (30-09-2021)

Identifiants

Citer

Piotr Antonik, Nicolas Marsal, Daniel Brunner, Damien Rontani. Bayesian Optimisation of Large-scale Photonic Reservoir Computers. Cognitive Computation, 2021, pp.1-9. ⟨10.1007/s12559-020-09732-6⟩. ⟨hal-03360041⟩
133 Consultations
124 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More