“…The GPGPU APIs have simplified the development of neural algorithms and ANNs for the graphics hardware significantly [10,16] and a variety of neurocomputing algorithms were ported to the GPUs [10,14,15,16,18,20,22,24]. The CUDA platform was used to achieve 46 to 63 times faster learning of a feedforward ANN by the backpropagation algorithm by Sierra-Canto et al [24] while Lopes and Ribeiro [14] reported a 10 to 40 faster implementation of the multiple backpropagation training of feedforward and multiple feedforward ANNs.…”
Section: Gpu Computingmentioning
confidence: 99%
“…Ghuzva et al [10] presented a coarse-grained implementation of the multilayer perceptron (MLP) on the CUDA platform that operated a set of MLPs in parallel 50 times faster than a sequential CPU-based implementation. The training of a feedforward neural network by genetic algorithms was implemented on CUDA by Patulea et al [20] and it was 10 times faster than a sequential version of the same algorithm. An application of a GPU-powered ANN for speech recognition is due to Scanzio et al [22].…”
Section: Gpu Computingmentioning
confidence: 99%
“…Modern many-core GPUs have been successfully used to accelerate a variety of meta-heuristics and bio-inspired algorithms [6,12,13] including different types of artificial neural networks [1,10,11,14,15,17,18,20,22,24]. To fully utilize the parallel hardware, the algorithms have to be carefully adapted to data-parallel architecture of the GPUs [21].…”
This study introduces a novel fine-grained parallel implementation of a neural principal component analysis (neural PCA) variant and the maximum Likelihood Hebbian Learning (MLHL) network designed for modern many-core graphics processing units (GPUs). The parallel implementation as well as the computational experiments conducted in order to evaluate the speedup achieved by the GPU are presented and discussed. The evaluation was done on a well-known artificial data set, the 2D bars data set.
“…The GPGPU APIs have simplified the development of neural algorithms and ANNs for the graphics hardware significantly [10,16] and a variety of neurocomputing algorithms were ported to the GPUs [10,14,15,16,18,20,22,24]. The CUDA platform was used to achieve 46 to 63 times faster learning of a feedforward ANN by the backpropagation algorithm by Sierra-Canto et al [24] while Lopes and Ribeiro [14] reported a 10 to 40 faster implementation of the multiple backpropagation training of feedforward and multiple feedforward ANNs.…”
Section: Gpu Computingmentioning
confidence: 99%
“…Ghuzva et al [10] presented a coarse-grained implementation of the multilayer perceptron (MLP) on the CUDA platform that operated a set of MLPs in parallel 50 times faster than a sequential CPU-based implementation. The training of a feedforward neural network by genetic algorithms was implemented on CUDA by Patulea et al [20] and it was 10 times faster than a sequential version of the same algorithm. An application of a GPU-powered ANN for speech recognition is due to Scanzio et al [22].…”
Section: Gpu Computingmentioning
confidence: 99%
“…Modern many-core GPUs have been successfully used to accelerate a variety of meta-heuristics and bio-inspired algorithms [6,12,13] including different types of artificial neural networks [1,10,11,14,15,17,18,20,22,24]. To fully utilize the parallel hardware, the algorithms have to be carefully adapted to data-parallel architecture of the GPUs [21].…”
This study introduces a novel fine-grained parallel implementation of a neural principal component analysis (neural PCA) variant and the maximum Likelihood Hebbian Learning (MLHL) network designed for modern many-core graphics processing units (GPUs). The parallel implementation as well as the computational experiments conducted in order to evaluate the speedup achieved by the GPU are presented and discussed. The evaluation was done on a well-known artificial data set, the 2D bars data set.
“…The training of EA-ANN on the graphic platform accelerator is ideal, due to the features of NVIDIA's compute unified device architecture (CUDA) which is a programming framework for the massively parallel GPUs, also to the architecture of GPU that contains thousands of independent floating-point units connected to on-board memory enabling high memory bandwidth, making this device perfect for providing significant speed-up to the intensive parallel applications (Patulea et al, 2014), by respecting memory access rules.…”
This paper proposes a novel parallel hybrid training approach to conceive an evolutionary robot. The proposed design aims to provide efficient behaviours to perform its tasks in a complex area such as walking toward a hidden destination. Embedded in robot brain, this training and evolution combination is typically accomplished by evolving considerable recurrent neural networks (RNNs) using an evolutionary strategy (ES). The effectiveness of this proposal is improved by employing CUDA technology that executes the evolutionary process of RNNs in a parallel way. The modifications applied are indicating to meet CUDA requirements in terms of CPU/GPU cooperation and memory management. Using a set of experiments performed by GPGPU-based physical simulator named open dynamics engine (ODE) and CUDA-based evolution, the effectiveness of the proposed parallel evolutionary training technique was validated for real movements of humanoid robots. This validation showed a promising speed-up, since this field requires very high powerful computational resources.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.