Frank Br��ckerhoff-Pl��ckelmann, Ivonne Bente, Marlon Becker, Niklas Vollmar, Nikolaos Farmakidis, Emma Lomonte, Francesco Lenzini, C David Wright, Harish Bhaskaran, Martin Salinga, Benjamin Risse, Wolfram H P Pernice
Author Information
Frank Br��ckerhoff-Pl��ckelmann: Physical Institute, University of M��nster, Heisenbergstra��e 11, 48149 M��nster, Germany. ORCID
Ivonne Bente: Physical Institute, University of M��nster, Heisenbergstra��e 11, 48149 M��nster, Germany. ORCID
Marlon Becker: Institute for Geoinformatics, University of M��nster, Heisenbergstra��e 2, 48149 M��nster, Germany.
Niklas Vollmar: Institute of Materials Physics, University of M��nster, Wilhelm-Klemm-Stra��e 10, 48149 M��nster, Germany.
Nikolaos Farmakidis: Department of Material, University of Oxford, Parks Road, Oxford OX1 3PH, UK. ORCID
Emma Lomonte: Physical Institute, University of M��nster, Heisenbergstra��e 11, 48149 M��nster, Germany. ORCID
Francesco Lenzini: Physical Institute, University of M��nster, Heisenbergstra��e 11, 48149 M��nster, Germany. ORCID
C David Wright: Department of Engineering, University of Exeter, North Park Road, Exeter EX4 4QF, UK. ORCID
Harish Bhaskaran: Department of Material, University of Oxford, Parks Road, Oxford OX1 3PH, UK. ORCID
Martin Salinga: Institute of Materials Physics, University of M��nster, Wilhelm-Klemm-Stra��e 10, 48149 M��nster, Germany. ORCID
Benjamin Risse: Institute for Geoinformatics, University of M��nster, Heisenbergstra��e 2, 48149 M��nster, Germany. ORCID
Wolfram H P Pernice: Physical Institute, University of M��nster, Heisenbergstra��e 11, 48149 M��nster, Germany. ORCID
We present an adaptive optical neural network based on a large-scale event-driven architecture. In addition to changing the synaptic weights (synaptic plasticity), the optical neural network's structure can also be reconfigured enabling various functionalities (structural plasticity). Key building blocks are wavelength-addressable artificial neurons with embedded phase-change materials that implement nonlinear activation functions and nonvolatile memory. Using multimode focusing, the activation function features both excitatory and inhibitory responses and shows a reversible switching contrast of 3.2 decibels. We train the neural network to distinguish between English and German text samples via an evolutionary algorithm. We investigate both the synaptic and structural plasticity during the training process. On the basis of this concept, we realize a large-scale network consisting of 736 subnetworks with 16 phase-change material neurons each. Overall, 8398 neurons are functional, highlighting the scalability of the photonic architecture.
References
Opt Lett. 2017 Apr 15;42(8):1560-1563
[PMID: 28409798]
