Timezone: »
The recent progress on automatically searching augmentation policies has boosted the performance substantially for various tasks. A key component of automatic augmentation search is the evaluation process for a particular augmentation policy, which is utilized to return reward and usually runs thousands of times. A plain evaluation process, which includes full model training and validation, would be time-consuming. To achieve efficiency, many choose to sacrifice evaluation reliability for speed. In this paper, we dive into the dynamics of augmented training of the model. This inspires us to design a powerful and efficient proxy task based on the Augmentation-Wise Weight Sharing (AWS) to form a fast yet accurate evaluation process in an elegant way. Comprehensive analysis verifies the superiority of this approach in terms of effectiveness and efficiency. The augmentation policies found by our method achieve superior accuracies compared with existing auto-augmentation search methods. On CIFAR-10, we achieve a top-1 error rate of 1.24%, which is currently the best performing single model without extra training data. On ImageNet, we get a top-1 error rate of 20.36% for ResNet-50, which leads to 3.34% absolute error rate reduction over the baseline augmentation.
Author Information
Keyu Tian (Sensetime; Beihang University)
Chen Lin (University of Oxford)
Ming Sun (SenseTime Group Limited)
Luping Zhou (University of Sydney)
Junjie Yan (Sensetime Group Limited)
Wanli Ouyang (The University of Sydney)
More from the Same Authors
-
2021 : MQBench: Towards Reproducible and Deployable Model Quantization Benchmark »
Yuhang Li · Mingzhu Shen · Jian Ma · Yan Ren · Mingxin Zhao · Qi Zhang · Ruihao Gong · Fengwei Yu · Junjie Yan -
2022 Poster: Improving Barely Supervised Learning by Discriminating Unlabeled Samples with Super-Class »
Guan Gui · Zhen Zhao · Lei Qi · Luping Zhou · Lei Wang · Yinghuan Shi -
2022 Spotlight: Improving Barely Supervised Learning by Discriminating Unlabeled Samples with Super-Class »
Guan Gui · Zhen Zhao · Lei Qi · Luping Zhou · Lei Wang · Yinghuan Shi -
2022 Spotlight: Lightning Talks 3A-1 »
Shu Ding · Wanxing Chang · Jiyang Guan · Mouxiang Chen · Guan Gui · Yue Tan · Shiyun Lin · Guodong Long · Yuze Han · Wei Wang · Zhen Zhao · Ye Shi · Jian Liang · Chenghao Liu · Lei Qi · Ran He · Jie Ma · Zemin Liu · Xiang Li · Hoang Tuan · Luping Zhou · Zhihua Zhang · Jianling Sun · Jingya Wang · LU LIU · Tianyi Zhou · Lei Wang · Jing Jiang · Yinghuan Shi -
2021 Poster: A Continuous Mapping For Augmentation Design »
Keyu Tian · Chen Lin · Ser Nam Lim · Wanli Ouyang · Puneet Dokania · Philip Torr -
2019 Poster: Efficient Neural Architecture Transformation Search in Channel-Level for Object Detection »
Junran Peng · Ming Sun · ZHAO-XIANG ZHANG · Tieniu Tan · Junjie Yan -
2018 Poster: FishNet: A Versatile Backbone for Image, Region, and Pixel Level Prediction »
Shuyang Sun · Jiangmiao Pang · Jianping Shi · Shuai Yi · Wanli Ouyang -
2018 Poster: Synaptic Strength For Convolutional Neural Network »
CHEN LIN · Zhao Zhong · Wu Wei · Junjie Yan