Title :
GPU-based simulation of spiking neural networks with real-time performance & high accuracy
Author :
Yudanov, Dmitri ; Shaaban, Muhammad ; Melton, Roy ; Reznik, Leon
Author_Institution :
Dept. of Comput. Eng., Rochester Inst. of Technol., Rochester, NY, USA
Abstract :
A novel GPU-based simulation of spiking neural networks is implemented as a hybrid system using Parker-Sochacki numerical integration method with adaptive order. Full single-precision floating-point accuracy for all model variables is achieved. The implementation is validated with exact matching of all neuron potential traces from GPU-based simulation versus those of a reference CPU-based simulation. A network of 4096 Izhikevich neurons simulated on an NVIDIA GTX260 device achieves real-time performance with a speedup of 9 compared to simulation executed on Opteron 285, 2.6-GHz device.
Keywords :
computer graphic equipment; coprocessors; digital simulation; neural nets; real-time systems; 4096 Izhikevich neurons; CPU-based simulation; GPU-based simulation; NVIDIA GTX260 device; Parker-Sochacki numerical integration method; adaptive order; hybrid system; neuron potential traces exact matching; real-time performance; single-precision floating-point accuracy; spiking neural networks; Accuracy; Computational modeling; Instruction sets; Kernel; Mathematical model; Neurons; Numerical models; CUDA; GPU; STDP; high accuracy; parallel computing; shared memory; spiking neural network;
Conference_Titel :
Neural Networks (IJCNN), The 2010 International Joint Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4244-6916-1
DOI :
10.1109/IJCNN.2010.5596334
