Going Deeper in Spiking Neural Networks: VGG and Residual Architectures

Abhronil Sengupta1, Yuting Ye2, Robert Wang2, Chiao Liu2, Kaushik Roy1
1Department of Electrical and Computer Engineering, Purdue University, United States
2Facebook Reality Labs, Facebook Research, United States

Tóm tắt

Từ khóa


Tài liệu tham khảo

Akopyan, 2015, TrueNorth: design and tool flow of a 65 mw 1 million neuron programmable neurosynaptic chip, IEEE Trans. Comp. Aided Design Integr. Circ. Syst., 34, 1537, 10.1109/TCAD.2015.2474396

Cao, 2015, Spiking deep convolutional neural networks for energy-efficient object recognition, Int. J. Comp. Vision, 113, 54, 10.1007/s11263-014-0788-3

Chen, 1998, Estimation of standby leakage power in CMOS circuit considering accurate modeling of transistor stacks, Proceedings, International Symposium on Low Power Electronics and Design, 1998, 239

Diehl, 2015, Unsupervised learning of digit recognition using spike-timing-dependent plasticity, Front. Comput. Neurosci., 9, 99, 10.3389/fncom.2015.00099

Diehl, 2015, Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing, 2015 International Joint Conference on Neural Networks (IJCNN), 1, 10.1109/IJCNN.2015.7280696

Esser, 2016, Convolutional networks for fast, energy-efficient neuromorphic computing, Proc. Natl. Acad. Sci. U.S.A., 113, 11441, 10.1073/pnas.1604850113

Farabet, 2012, Comparison between frame-constrained fix-pixel-value and frame-free spiking-dynamic-pixel ConvNets for visual processing, Front. Neurosci., 6, 32, 10.3389/fnins.2012.00032

Han, , Deep compression: compressing deep neural networks with pruning, trained quantization and huffman coding, arXiv preprint arXiv:1510.00149

Han, , Learning both weights and connections for efficient neural network, Advances in Neural Information Processing Systems, 1135

Hardt, 2016, Identity matters in deep learning, arXiv preprint arXiv:1611.04231

He, , Deep residual learning for image recognition, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770

He, , Identity mappings in deep residual networks, European Conference on Computer Vision, 630

Hunsberger, 2016, Training spiking deep networks for neuromorphic hardware, arXiv preprint arXiv:1611.05141

Ioffe, 2015, Batch normalization: accelerating deep network training by reducing internal covariate shift, International Conference on Machine Learning, 448

Izhikevich, 2003, Simple model of spiking neurons, IEEE Trans. Neural Netw., 14, 1569, 10.1109/TNN.2003.820440

KrizhevskyA. HintonG. Learning multiple layers of features from tiny images2009

Krizhevsky, 2012, ImageNet classification with deep convolutional neural networks, Advances in Neural Information Processing Systems, 1097

LeCun, 1998, Gradient-based learning applied to document recognition, Proceed. IEEE, 86, 2278, 10.1109/5.726791

Merolla, 2014, A million spiking-neuron integrated circuit with a scalable communication network and interface, Science, 345, 668, 10.1126/science.1254642

Pérez-Carrasco, 2013, Mapping from frame-driven to frame-free event-driven vision systems by low-Rate rate coding and coincidence processing–application to feedforward convNets, IEEE Trans. Pattern Anal. Mach. Intelligence, 35, 2706, 10.1109/TPAMI.2013.71

Posch, 2011, A QVGA 143 dB dynamic range frame-free PWM image sensor with lossless pixel-level video compression and time-domain CDS, IEEE J. Solid-State Circ., 46, 259, 10.1109/JSSC.2010.2085952

Posch, 2014, Retinomorphic event-based vision sensors: bioinspired cameras with spiking output, Proc. IEEE, 102, 1470, 10.1109/JPROC.2014.2346153

Rueckauer, 2017, Conversion of continuous-valued deep networks to efficient event-driven networks for image classification, Front. Neurosci., 11, 682, 10.3389/fnins.2017.00682

Russakovsky, 2015, ImageNet large scale visual recognition challenge, Int. J. Comp. Visi., 115, 211, 10.1007/s11263-015-0816-y

Simonyan, 2014, Very deep convolutional networks for large-scale image recognition, arXiv preprint arXiv:1409.1556

Srivastava, 2014, Dropout: A simple way to prevent neural networks from overfitting, J. Mach. Learn. Res., 15, 1929

Zhao, 2015, Feedforward categorization on AER motion events using cortex-like features in a spiking neural network, IEEE Trans. Neural Netw. Learn. Syst., 26, 1963, 10.1109/TNNLS.2014.2362542