Computing with Spiking Neuron Networks

International audienceSpiking Neuron Networks (SNNs) are often referred to as the 3rd generation ofneural networks. Highly inspired from natural computing in the brain andrecent advances in neurosciences, they derive their strength and interest from anaccurate modeling of synaptic interactions between neurons, taking into account thetime of spike firing. SNNs overcome the computational power of neural networksmade of threshold or sigmoidal units. Based on dynamic event-driven processing,they open up new horizons for developing models with an exponential capacity ofmemorizing and a strong ability to fast adaptation. Today, the main challenge is todiscover efficient learning rules that might take advantage of the specific featuresof SNNs while keeping the nice properties (general-purpose, easy-to-use, availablesimulators, etc.) of traditional connectionist models. This chapter relates the history of the “spiking neuron” in Section 1 and summarizes the most currently-in-usemodels of neurons and synaptic plasticity in Section 2. The computational power ofSNNs is addressed in Section 3 and the problem of learning in networks of spikingneurons is tackled in Section 4, with insights into the tracks currently explored forsolving it. Finally, Section 5 discusses application domains, implementation issuesand proposes several simulation frameworks