Simulation of spiking neural networks — architectures and implementations

Abstract

The fast simulation of large networks of spiking neurons is a major task for the examination of biology-inspired vision systems. Networks of this type label features by synchronization of spikes and there is strong demand to simulate these e ects in real world environments. As the calculations for one model neuron are complex, the digital simulation of large networks is not e cient using existing simulation systems. Consequently, it is necessary to develop special simulation techniques. This article introduces a wide range of concepts for the di erent parts of digital simulator systems for large vision networks and presents accelerators based on these foundations. Pulse-Coded Neural Vision NetworksThe communication in PCNNs is based on spike e x c hange. In contrast to conventional model neurons, e.g. McCulloch & Pitts neurons, the generation of a spike requires high computational e ort in connection with the time behaviour in the biological example. The computational e ort for individual neuron calculations compared to whole network processing is much higher in PCNNs than in conventional ANNs.Common simulation techniques for neural networks make use of vector representations for the neurons and matrix representations for the connection network 11]. These techniques are not suitable for PCNNs because the actual activity of one neuron is not representable by o n l y o n e v alue. Hence, common simulation techniques based on the acceleration of matrix-vector-calculations are not su cient for PCNNs. A new simulation paradigm is required with

A real-time spiking cerebellum model for learning robot control

“…4. This neuron model is a modified version of the Spike-Response-Model (SRM) (Gerstner and Kistler, 2002) widely used in the literature (Eckhorn et al, 1990;Schoenauer et al, 2002;Shaefer et al, 2002) to study, for example, temporal coding issues (Eckhorn et al, 2004).…”

A GUI (Graphical User Interface) framework to Introduce H-H Model in Comparison with Kv3.3 Model

“…2002 Schaefer, M., et al [12] Gave the concept of digital simulation system. That required improved modeling in which rising phase of action potential can be modified.…”

Review of pulse-coupled neural networks

“…reducing the neuron calculations and the synapse calculations) to reduce the time of simulation. Remarkably, they combine commercial hardware with dedicated communication hardware to overcome this bottleneck in parallel processing [131]. Takahashi et al develop a simple circuit of the hyper-chaos generator, and this circuit can be used for the development of large scale PCNN hardware [132].…”