STDP-based spiking deep convolutional neural networks for object recognition. This is a good introduction to implementing spiking neural networks with unsupervised STDP-based learning for real-world tasks such as digit recognition. Unsupervised learning of digit recognition using spike-timing-dependent plasticity. Competitive STDP-based spike pattern learning. The MNIST Database of Handwritten Digits (1998).
Unsupervised regenerative learning of hierarchical features in spiking deep networks for object recognition. Supervised learning based on temporal coding in spiking neural networks. HFirst: a temporal approach to object recognition. Training deep spiking neural networks using backpropagation. NormAD: normalized approximate descent-based supervised learning rule for spiking neurons. A new supervised learning algorithm for multiple spiking neural networks with application in epilepsy and seizure detection. Error-backpropagation in temporally encoded networks of spiking neurons. The tempotron: a neuron that learns spike-timing-based decisions. Supervised learning in spiking neural networks with ReSuMe: sequence learning, classification, and spike shifting. Deep learning with spiking neurons: opportunities and challenges. Operating Systems Design and Implementation 265–283 (2016). Tensorflow: a system for large-scale machine learning. on Rebooting Computing 20 (IEEE, 2016).Ībadi, M. Conversion of artificial recurrent neural networks to spiking neural networks for low-power neuromorphic hardware. Conversion of continuous-valued deep networks to efficient event-driven networks for image classification. Rueckauer, B., Lungu, I.-A., Hu, Y., Pfeiffer, M. Mapping from frame-driven to frame-free event-driven vision systems by low-rate rate coding and coincidence processing-application to feedforward ConvNets. Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing. Spiking deep convolutional neural networks for energy-efficient object recognition. This paper was the first to demonstrate the competitive performance of a conversion-based spiking neural network on ImageNet data for deep neural architectures.Ĭao, Y., Chen, Y. Going deeper in spiking neural networks: VGG and residual architectures. A dataset for visual navigation with neuromorphic methods. Vision meets robotics: the KITTI dataset. DVS benchmark datasets for object tracking, action recognition, and object recognition. Rate coding versus temporal order coding: what the retinal ganglion cells tell the visual cortex. on Computer Vision and Pattern Recognition 248–255 (IEEE, 2009). ImageNet: a large-scale hierarchical image database. Improving neural networks by preventing co-adaptation of feature detectors. E., Srivastava, N., Krizhevsky, A., Sutskever, I. A spiking neural network model of 3D perception for event-based neuromorphic stereo vision systems. Asynchronous event-based binocular stereo matching. Rogister, P., Benosman, R., Ieng, S.-H., Lichtsteiner, P. Exploring flow, factors, and outcomes of temporal event sequences with the outflow visualization. Asynchronous frameless event-based optical flow. A review of current neuromorphic approaches for vision, auditory, and olfactory sensors. 128×128 120 db 15 μs latency asynchronous temporal contrast vision sensor. The Organization of Behavior: A Neuropsychological Theory (Wiley, 1949).Ībbott, L. This seminal work proposed gradient-descent-based backpropagation as a learning method for neural networks. Learning representations by back-propagating errors. Rectified linear units improve restricted Boltzmann machines. A logical calculus of the ideas immanent in nervous activity. This paper was one of the first works to provide a rigorous mathematical analysis of the computational power of spiking neurons, categorizing them as the third generation of neural networks (after perceptron and sigmoidal neurons). Networks of spiking neurons: the third generation of neural network models. In 21st Asia and South Pacific Design Automation Conf.
Efficient embedded learning for IoT devices. Stochastic dynamics as a principle of brain function. This work-using deep convolutional networks-was the first to win the ImageNet challenge, fuelling the subsequent deep-learning revolution.ĭeco, G., Rolls, E. et al.) 1097–1105 (Neural Information Processing Systems Foundation, 2012). In Advances in Neural Information Processing Systems Vol. ImageNet classification with deep convolutional neural networks. Distributed hierarchical processing in the primate cerebral cortex.
The economy of brain network organization. Neural networks and neuroscience-inspired computer vision. Mastering the game of Go without human knowledge.
0 kommentar(er)
