AutoGrow: Automatic Layer Growing in Deep Convolutional Networks

June 07, 2019 · Entered Twilight · 🏛 Knowledge Discovery and Data Mining

"Last commit was 6.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .gitignore, README.md, cifar10, imagenet

Authors Wei Wen, Feng Yan, Yiran Chen, Hai Li arXiv ID 1906.02909 Category cs.LG: Machine Learning Cross-listed cs.CV, cs.NE, stat.ML Citations 40 Venue Knowledge Discovery and Data Mining Repository https://github.com/wenwei202/autogrow ⭐ 40 Last Checked 1 month ago

Abstract

Depth is a key component of Deep Neural Networks (DNNs), however, designing depth is heuristic and requires many human efforts. We propose AutoGrow to automate depth discovery in DNNs: starting from a shallow seed architecture, AutoGrow grows new layers if the growth improves the accuracy; otherwise, stops growing and thus discovers the depth. We propose robust growing and stopping policies to generalize to different network architectures and datasets. Our experiments show that by applying the same policy to different network architectures, AutoGrow can always discover near-optimal depth on various datasets of MNIST, FashionMNIST, SVHN, CIFAR10, CIFAR100 and ImageNet. For example, in terms of accuracy-computation trade-off, AutoGrow discovers a better depth combination in ResNets than human experts. Our AutoGrow is efficient. It discovers depth within similar time of training a single DNN. Our code is available at https://github.com/wenwei202/autogrow.