• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Artificial Neural Networks >Automatic Augmentation by Hill Climbing
【24h】

Automatic Augmentation by Hill Climbing

机译：通过爬山自动增强

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

When learning from images, it is desirable to augment the dataset with plausible transformations of its images. Unfortunately, it is not always intuitive for the user how much shear or translation to apply. For this reason, training multiple models through hyperparameter search is required to find the best augmentation policies. But these methods are computationally expensive. Furthermore, since they generate static policies, they do not take advantage of smoothly introducing more aggressive augmentation transformations. In this work, we propose repeating each epoch twice with a small difference in data augmentation intensity, walking towards the best policy. This process doubles the number of epochs, but avoids having to train multiple models. The method is compared against random and Bayesian search for classification and segmentation tasks. The proposal improved twice over random search and was on par with Bayesian search for 4% of the training epochs.
机译：当从图像中学习时，期望用其图像的合理变换来扩充数据集。不幸的是，对于用户来说施加多少剪切或平移并不总是直观的。因此，需要通过超参数搜索来训练多个模型，以找到最佳的扩充策略。但是这些方法在计算上是昂贵的。此外，由于它们生成静态策略，因此无法利用平稳引入更具攻击性的增强转换的优势。在这项工作中，我们建议将每个时期重复两次，但数据增强强度的差异很小，朝着最佳策略迈进。此过程使时期数加倍，但避免了必须训练多个模型。将该方法与随机和贝叶斯搜索进行比较，以进行分类和分割任务。该建议比随机搜索改进了两次，并且与贝叶斯搜索在训练时期的4％方面相提并论。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号