Delays in photonic reservoir computing with semiconductor optical amplifiers
Kristof Vandoorne, David Verstraeten, Benjamin Schrauwen, Joni Dambre, Peter Bienstman · Ghent University Academic Bibliography (Ghent University) · 2010
Reservoir computing is a decade old framework from the field of machine learning to use and train recurrent neural networks and it splits the network in a reservoir that does the 'computation' and a simple readout function.This technique has been among the stateof-the-art for many classification problems.So far implementations have been mainly software based, but a hardware implementation offers the promise of being low-power and fast.We will show that photonic reservoirs can achieve an excellent performance on a benchmark isolated digit recognition task, if the interconnection delay is optimized and the phase controlled.Furthermore we will show that this optimal delay is dependent on the input speed of the audio signal.