Spatio-temporal pattern recognizers using spiking neurons and spike-timing-dependent plasticity

Abstract

It has previously been shown that by using spike-timing-dependent plasticity (STDP), neurons can adapt to the beginning of a repeating spatio-temporal firing pattern in their input. In the present work, we demonstrate that this mechanism can be extended to train recognizers for longer spatio-temporal input signals. Using a number of neurons that are mutually connected by plastic synapses and subject to a global winner-takes-all mechanism, chains of neurons can form where each neuron is selective to a different segment of a repeating input pattern, and the neurons are feed-forwardly connected in such a way that both the correct input segment and the firing of the previous neurons are required in order to activate the next neuron in the chain. This is akin to a simple class of finite state automata. We show that nearest-neighbor STDP (where only the pre-synaptic spike most recent to a post-synaptic one is considered) leads to “nearest-neighbor” chains where connections only form between subsequent states in a chain (similar to classic “synfire chains”). In contrast, “all-to-all spike-timing-dependent plasticity” (where all pre- and post-synaptic spike pairs matter) leads to multiple connections that can span several temporal stages in the chain; these connections respect the temporal order of the neurons. It is also demonstrated that previously learnt individual chains can be “stitched together” by repeatedly presenting them in a fixed order. This way longer sequence recognizers can be formed, and potentially also nested structures. Robustness of recognition with respect to speed variations in the input patterns is shown to depend on rise-times of post-synaptic potentials and the membrane noise. It is argued that the memory capacity of the model is high, but could theoretically be increased using sparse codes.

Optimal Localist and Distributed Coding of Spatiotemporal Spike Patterns Through STDP and Coincidence Detection

“…Surprisingly, a single neuron could robustly learn up to ~40 independent patterns (using parameters arguably in the biological range). This was not clear from previous simulations studies, in which neurons equipped with STDP only learned one pattern (localist coding) (Masquelier et al, 2008 , 2009 ; Gilson et al, 2011 ; Humble et al, 2012 ; Hunzinger et al, 2012 ; Kasabov et al, 2013 ; Klampfl and Maass, 2013 ; Nessler et al, 2013 ; Krunglevicius, 2015 ; Sun et al, 2016 ; Masquelier, 2017 ), or two patterns (Yger et al, 2015 ). This shows that STDP and coincidence detection are compatible with distributed coding.…”

NeuroFlow: A General Purpose Spiking Neural Network Simulation Platform using Customizable Processors

“…It demonstrates that conventional multi-core processors can be used along with the FPGA during a neural simulation in this system. Nearest-neighbor STDP is a common plasticity algorithm for learning in SNNs (Benuskova and Abraham, 2007 ; Babadi and Abbott, 2010 ; Humble et al, 2012 ). It is an efficient and good approximation of the STDP with all-to-all spike interaction as shown in Figure 5 .…”

Neuromorphic computing with resistive switching memory devices

“…This is opposite to how the brain represents information, which is mapped by the specific neuron being active, and by the specific time of the neuron action, or spike. Such a spatiotemporal coding is what makes the brain highly energy efficient and highly functional in representing and elaborating complex information [69][70][71]. Another conjectured mode of information coding in the brain is rate coding [69][70][71], where the average rate of neuron spiking is used to describe the relevant input/output information.…”