Delay learning and polychronization for reservoir computing

Abstract

We propose a multi-timescale learning rule for spiking neuron networks, in the line of the recently emerging field of reservoir computing. The reservoir is a network model of spiking neurons, with random topology and driven by STDP (spike-time-dependent plasticity), a temporal Hebbian unsupervised learning mode, biologically observed. The model is further driven by a supervised learning algorithm, based on a margin criterion, that affects the synaptic delays linking the network to the readout neurons, with classification as a goal task. The network processing and the resulting performance can be explained by the concept of polychronization, proposed by Izhikevich [Polychronization: computation with spikes, Neural Comput. 18(2) (2006) 245-282], on physiological grounds. The model emphasizes that polychronization can be used as a tool for exploiting the computational power of synaptic delays and for monitoring the topology and activity of a spiking neuron network. r

Computing with Spiking Neuron Networks

“…Another approach, by Paugam-Moisy et al [127], takes advantage of the theoretical results proving the importance of delays in computing with spiking neurons (see Section 3) for defining a supervised learning rule acting on the delays of connections (instead of weights) between the reservoir and the readout neurons. The reservoir is an SNN, with an STDP rule for adapting the weights to the task at hand, where can be observed that polychronous groups (see Section 3.2) are activated more and more selectively as training goes on.…”

Supervised Associative Learning in Spiking Neural Network

“…Learning is implemented in a Hebbian paradigm, considering both spike rate and timings of both pre-synaptic and post-synaptic neurons in a learning window [9]. In a learning trial with 500 milliseconds (ms) simulated time, the time window is divided into 100 ms (T=100) wide overlapping bins at 50 ms intervals (Fig.…”

“…The weight adjustments, ∆W are calculated as a function of time difference, ∆t = t j (f) -t i (f) , where t j (f) and t i (f) are the last firing times of post-synaptic neuron j and pre-synaptic neuron i, respectively, within the learning time bin (Fig. 2) [9]. To avoid saturation of synaptic strength values infinitely, we keep the values within the range 0 to 3.…”

“…In this study, we explore a supervised associative learning algorithm as a combination of spike emission rate dependent and STDP approaches from [9] on a learning task similar to [8]. Supervision in learning is only through intensified currents into paired target neuron subpopulations.…”

Computação por assembleias neurais em redes neurais pulsadas.

“…Redes de neurônios podem também processar, representar, ou codificar estímulos por meio de cadeias de neurônios pulsando sincronamente, o que traz a noção de assembleias de neurônios [47], [80], [81], [44].…”