Competitive residual neural network for image classification

Ren, 2017, Faster r-cnn: Towards real-time object detection with region proposal networks, IEEE Trans. Pattern Anal. Mach. Intell., 39, 1137, 10.1109/TPAMI.2016.2577031 Rawat, 2017, Deep convolutional neural networks for image classification: A comprehensive review, Neural Comput., 29, 2352, 10.1162/neco_a_00990 Goodfellow, 2016 Pang, 2017, Deep learning and preference learning for object tracking: A combined approach, Neural Process. Lett., 1 LeCun, 1990, Handwritten digit recognition with a back-propagation network, 396 LeCun, 2015, Deep learning, Nature, 521, 436, 10.1038/nature14539 Krizhevsky, 2012, Imagenet classification with deep convolutional neural networks, 1097 Simonyan, 2015, Very deep convolutional networks for large-scale image recognition Szegedy, 2015, Going deeper with convolutions, 1 Szegedy, 2016, Rethinking the inception architecture for computer vision, 2818 He, 2016, Deep residual learning for image recognition, 770 Glorot, 2010, Understanding the difficulty of training deep feedforward neural networks, 249 He, 2015, Delving deep into rectifiers: Surpassing human-level performance on imagenet classification, 1026 Bengio, 2012, 437 Ioffe, 2015, Batch normalization: Accelerating deep network training by reducing internal covariate shift, 448 Huang, 2017, Densely connected convolutional networks, 4700 Nair, 2010, Rectified linear units improve restricted boltzmann machines, 807 He, 2016, Identity mappings in deep residual networks, 630 Zagoruyko, 2016, Wide residual networks, 1 L. Zhao, J. Wang, X. Li, Z. Tu, W. Zeng, On the connection of deep fusion to ensembling, Tech. rep. (2016). URL https://www.microsoft.com/en-us/research/publication/connection-deep-fusion-ensembling/. Goodfellow, 2013, Maxout networks, III Liao, 2017, A deep convolutional neural network module that promotes competition of multiple-size filters, Pattern Recognit., 71, 94, 10.1016/j.patcog.2017.05.024 A. Krizhevsky, V. Nair, G. Hinton, The cifar-10 and cifar-100 datasets, online: http://www.cs.toronto.edu/kriz/cifar.html. Netzer, 2011, Reading digits in natural images with unsupervised feature learning, 5 Russakovsky, 2015, Imagenet large scale visual recognition challenge, Int. J. Comput. Vis., 115, 211, 10.1007/s11263-015-0816-y Lin, 2014, Microsoft coco: Common objects in context, 740 Everingham, 2010, The pascal visual object classes (voc) challenge, Int. J. Comput. Vis., 88, 303, 10.1007/s11263-009-0275-4 G.E. Hinton, N. Srivastava, A. Krizhevsky, I. Sutskever, R.R. Salakhutdinov, Improving neural networks by preventing co-adaptation of feature detectors (2012). arXiv:1207.0580. Xie, 2017, Aggregated residual transformations for deep neural networks, 5987 Srivastava, 2015, Highway networks Larsson, 2017, Fractalnet: Ultra-deep neural networks without residuals Lin, 2013, Network in network Y. LeCun, C. Cortes, C.J. Burges, The mnist database, online: http://yann.lecun.com/exdb/mnist/. Maas, 2013, Rectifier nonlinearities improve neural network acoustic models Clevert, 2016, Fast and accurate deep network learning by exponential linear units (elus) Srivastava, 2015, Understanding locally competitive networks Liao, 2016, On the importance of normalisation layers in deep learning with piecewise linear activation units, 1 Nesterov, 1983, A method of solving a convex programming problem with convergence rate o (1/k2), 372 P. Goyal, P. Dollár, R.B. Girshick, P. Noordhuis, L. Wesolowski, A. Kyrola, A. Tulloch, Y. Jia, K. He, Accurate, large minibatch SGD: training imagenet in 1 hour, Tech. rep. Facebook (2017). URL https://research.fb.com/publications/accurate-large-minibatch-sgd-training-imagenet-in-1-hour/. Springenberg, 2015, Striving for simplicity: The all convolutional net Lee, 2015, Deeply-supervised nets, 562