论文信息 - An FPGA-based model suitable for evolution and development of spiking neural networks

An FPGA-based model suitable for evolution and development of spiking neural networks

We propose a digital neuron model suitable for evolving and growing heterogeneous spiking neural networks on FPGAs using a piecewise linear approximation of the Quadratic Integrate and Fire (QIF) model. A network of 161 neurons and 1610 synapses with 4210 times real- time neuron simulation speed was simulated and synthesized for a Virtex-5 chip.

[1] Dario Floreano,et al. Evolutionary morphogenesis for multi-cellular systems , 2007, Genetic Programming and Evolvable Machines.

[2] Eugene M. Izhikevich,et al. Dynamical Systems in Neuroscience: The Geometry of Excitability and Bursting , 2006 .

[3] Andres Upegui,et al. An FPGA platform for on-line topology exploration of spiking neural networks , 2005, Microprocess. Microsystems.

[4] Benjamin Schrauwen,et al. Parallel hardware implementation of a broad class of spiking neurons using serial arithmetic , 2006, ESANN.

[5] Andres Upegui,et al. Evolving Hardware with Self-reconfigurable connectivity in Xilinx FPGAs , 2006, First NASA/ESA Conference on Adaptive Hardware and Systems (AHS'06).

[6] Eduardo Sanchez,et al. The POEtic Electronic Tissue and Its Role in the Emulation of Large-Scale Biologically Inspired Spiking Neural Networks Models , 2006, Complexus.