GPU-based simulation of spiking neural networks with real-time performance &amp;amp; high accuracy

Yudanov, Dmitri; Shaaban, Muhammad; Melton, Rachel E.; Reznik, Leon

doi:10.1109/ijcnn.2010.5596334

Cited by 38 publications

(21 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, he compares these parameters between a GPU and a FPGA programming Gilbert' algorithm. Yudanov [13], implements a hybrid method with numeric integration of Parker Sochacki (PS) with adaptative order. This is validated at the moment in the comparision made between GPU and CPU in their characteristics.…”

Section: Relating Work On the Use Of Parallel Programming In Artifmentioning

confidence: 99%

“…This describes propagation and generation of potential of a big axon of squid in order to explain the main properties. SNN's is the model most similar to the neurons of mammals [27].SNN can be applied to the same problems that depend on behavior of time of parameters because of its singular characteristic of coding in the time [13]. In case of SVM, in decade of 1990's was started the development [34], even this methodology had been invented since 1979 [31] by Vapnik, to solve more complex problems,linearlly separable or non -separable.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Spiking Neural Networks and Support Vector Machine Performance Byusinggpu's

Tabarez-Paz¹,

Hernández-Gress²,

González-Mendoza³

2013

IJSC

View full text Add to dashboard Cite

show abstract

Section: Relating Work On the Use Of Parallel Programming In Artifmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Survey of Spiking Neural Networks and Support Vector Machine Performance Byusinggpu's

Tabarez-Paz¹,

Hernández-Gress²,

González-Mendoza³

2013

IJSC

View full text Add to dashboard Cite

show abstract

“…These enhancements, in combination with parallel computing (Bower and Beeman, 1998; Migliore et al, 2006), have become a necessity to cope with the higher computational and the communication demands of neuroapplications. Recently, a number of developers have investigated the possibility of simulating spiking neural networks on a single Graphical Processing Unit (GPU) (Bernhard and Keriven, 2005; Fernandez et al, 2008; Fidjeland et al, 2009; Nageswaran et al, 2009a,b; Tiesel and Maida, 2009; Bhuiyan et al, 2010; Fidjeland and Shanahan, 2010; Han and Taha, 2010a,b; Hoffmann et al, 2010; Mutch et al, 2010; Scorciono, 2010; Yudanov et al, 2010; Nowotny, 2011; Ahmadi and Soleimani, 2011; Igarashi et al, 2011; Thibeault et al, 2011; Wang et al, 2011) or on multiple Graphics Processing Units (GPUs) (Brette and Goodman, 2012b). All these current simulators have shown significant improvements over their CPU only counterparts by integrating the utilization of GPUs.…”

Section: Introductionmentioning

confidence: 99%

A novel CPU/GPU simulation environment for large-scale biologically realistic neural modeling

Hoang¹,

Tanna²,

Bray³

et al. 2013

Front. Neuroinform.

View full text Add to dashboard Cite

Computational Neuroscience is an emerging field that provides unique opportunities to study complex brain structures through realistic neural simulations. However, as biological details are added to models, the execution time for the simulation becomes longer. Graphics Processing Units (GPUs) are now being utilized to accelerate simulations due to their ability to perform computations in parallel. As such, they have shown significant improvement in execution time compared to Central Processing Units (CPUs). Most neural simulators utilize either multiple CPUs or a single GPU for better performance, but still show limitations in execution time when biological details are not sacrificed. Therefore, we present a novel CPU/GPU simulation environment for large-scale biological networks, the NeoCortical Simulator version 6 (NCS6). NCS6 is a free, open-source, parallelizable, and scalable simulator, designed to run on clusters of multiple machines, potentially with high performance computing devices in each of them. It has built-in leaky-integrate-and-fire (LIF) and Izhikevich (IZH) neuron models, but users also have the capability to design their own plug-in interface for different neuron types as desired. NCS6 is currently able to simulate one million cells and 100 million synapses in quasi real time by distributing data across eight machines with each having two video cards.

show abstract

“…A Geometrical Support Vector Machine classifier has also been implemented using GPGPU [HWS10]. It extends different GPGPU implementations for Neural Networks [YSMR10]. Adaboost is also a widely applied classifier.…”

Section: Introductionmentioning

confidence: 99%

Intelligent GPGPU Classification in Volume Visualization: A framework based on Error‐Correcting Output Codes

Escalera¹,

Puig²,

Amoros³

et al. 2011

Computer Graphics Forum

View full text Add to dashboard Cite

In volume visualization, the definition of the regions of interest is inherently an iterative trial-and-error process finding out the best parameters to classify and render the final image. Generally, the user requires a lot of expertise to analyze and edit these parameters through multi-dimensional transfer functions. In this paper, we present a framework of intelligent methods to label on-demand multiple regions of interest. These methods can be split into a two-level GPU-based labelling algorithm that computes in time of rendering a set of labelled structures using the Machine Learning Error-Correcting Output Codes (ECOC) framework. In a pre-processing step, ECOC trains a set of Adaboost binary classifiers from a reduced pre-labelled data set. Then, at the testing stage, each classifier is independently applied on the features of a set of unlabelled samples and combined to perform multi-class labelling. We also propose an alternative representation of these classifiers that allows to highly parallelize the testing stage. To exploit that parallelism we implemented the testing stage in GPU-OpenCL. The empirical results on different data sets for several volume structures shows high computational performance and classification accuracy.

show abstract

GPU-based simulation of spiking neural networks with real-time performance & high accuracy

Cited by 38 publications

References 12 publications

A Survey of Spiking Neural Networks and Support Vector Machine Performance Byusinggpu's

A Survey of Spiking Neural Networks and Support Vector Machine Performance Byusinggpu's

A novel CPU/GPU simulation environment for large-scale biologically realistic neural modeling

Intelligent GPGPU Classification in Volume Visualization: A framework based on Error‐Correcting Output Codes

Contact Info

Product

Resources

About