Efficient, Scalable and Parallel Event-Driven Simulation Techniques for Complex Spiking Neuron Models

The simulation of spiking neural networks (SNNs), both for neuro-biological and application based simulations, is known to be a very time consuming task. This limits the size of SNN that can be simulated in reasonable time or forces users to overly limit the complexity of the neuron models. This is one of the driving forces behind much of the recent research on event-driven simulation strategies. Although event-driven simulation allows for precise and efficient simulation of certain spiking neuron models, it is not straightforward to generalize the technique to more complex neuron models, mostly because the firing time of these neuron models is computationally expensive to eval-uate. Most solutions proposed in literature concentrate on algorithms that can solve this evaluation efficiently. How-ever, these solutions do not scale well when more state variables are involved in the neuron model, which is, for example, the case when multiple synaptic time constants are used. In more complex neuron models, e.g. conductance based neuron models, it is usually the case that the dy-namics of the membrane potential cannot be solved an-alytically. This complicates the implementation of such models in an event-driven framework in order to benefit from the potentially high efficiency and accuracy of the event-driven simulation technique