Browsing by Author "Sackesyn, Stijn"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Improving Time Series Recognition and Prediction with Networks and Ensembles of Passive Photonic Reservoirs(IEEE Journal of Selected Topics in Quantum Electronics, 2019) Freiberger, Matthias; Sackesyn, Stijn; Ma, Chonghuai; Katumba, Andrew; Bienstman, Peter; Dambre, JoniAs the performance increase of traditional Von- Neumann computing attenuates, new approaches to computing need to be found. A promising approach for low-power computing at high bitrates is integrated photonic reservoir computing. In the past though, the feasible reservoir size and computational power of integrated photonic reservoirs have been limited by hardware constraints. An alternative solution to building larger reservoirs is the combination of several small reservoirs to match or exceed the performance of a single bigger one. This work summarizes our efforts to increase the available computational power by combining multiple reservoirs into a single computing architecture. We investigate several possible combination techniques and evaluate their performance using the classic XOR and header recognition tasks as well as the well-known Santa Fe chaotic laser prediction task. Our findings suggest that a new paradigm of feeding a reservoir’s output into the readout structure of the next one shows consistently good results for various tasks as well as for both electrical and optical readouts and coupling schemes.Item Neuromorphic computing based on silicon photonics(IEEE Journal of Selected Topics in Quantum Electronics, 2018) Katumba, Andrew; Freiberger, Matthias; Laporte, Floris; Lugnan, Alessio; Sackesyn, Stijn; Ma, Chonghuai; Dambre, Joni; Bienstman, PeterWe present our latest progress using new neuromorphic paradigms for optical information processing in silicon photonics. We show how passive reservoir computing chips can be used to perform a variety of tasks (bit level tasks, nonlinear dispersion compensation, ...) at high speeds and low power consumption. In addition, we present a spatial analog of reservoir computing based on pillar scatterers and a cavity, that can be used to speed up classification of biological cells.Item A power-efficient architecture for silicon photonic reservoir computing(EU project PHRESCO, 2009) Sackesyn, Stijn; Ma, Chonghuai; Katumba, Andrew; Denis-le Coarer, Florian; Dambre, Joni; Bienstman, PeterReservoir computing is a machine learning technique in which a nonlinear dynamical system which is also called the reservoir is used for computation. While it was originally implemented as an efficient way to train a neural network [1], it has now grown to a method which is commonly used for classification and regression tasks. Unlike other techniques which optimize the recurrent neural network (RNN) itself to solve a task, the RNN is not modified during training in reservoir computing. Instead, a linear combination is used to train an optimal classifier in the high dynamical state space to which the signal is projected by propagation through the reservoir. By keeping the recurrent network unchanged and train only on the level of the linear output layer makes reservoir computing a computationally cheap method.