I NSTANCE –S PECIFIC A UGMENTATION : C APTURING L OCAL I NVARIANCES-猫眼课题宝

DOI：

发表时间：

2022

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Ning Miao;Tom Rainforth;Emile Mathieu;Yann Dubois;Y. Teh;Adam Foster;Hyunjik Kim

我们介绍InstaAug，一种从数据中自动学习特定输入增强的方法。以前的数据增强方法通常假设原始输入和应用于该输入的变换之间是独立的。这可能是高度限制性的，因为增强所基于的不变性本身通常是高度依赖于输入的;例如，我们可以把一片叶子从绿色变成黄色，同时保持它的标签，但不能把一个酸橙。InstaAug通过引入一个将输入映射到定制的转换分布的不变性模块来允许输入依赖性。它可以以完全端到端的方式与下游模型一起同时训练，或者为预训练模型单独学习。我们的经验表明，InstaAug学习有意义的输入依赖增强广泛的转换类，这反过来又提供了更好的性能监督和自我监督的任务。代码可在https://github.com/NingMiao/InstaAug上获得。

We introduce InstaAug, a method for automatically learning input-specific augmentations from data. Previous data augmentation methods have generally assumed independence between the original input and the transformation applied to that input. This can be highly restrictive, as the invariances that the augmentations are based on are themselves often highly input dependent; e.g., we can change a leaf from green to yellow while maintaining its label, but not a lime. InstaAug instead allows for input dependency by introducing an invariance module that maps inputs to tailored transformation distributions. It can be simultaneously trained alongside the downstream model in a fully end-to-end manner, or separately learned for a pre-trained model. We empirically demonstrate that InstaAug learns meaningful input-dependent augmentations for a wide range of transformation classes, which in turn provides better performance on both supervised and self-supervised tasks. Codes are available at https://github.com/NingMiao/InstaAug.

I NSTANCE –S PECIFIC A UGMENTATION : C APTURING L OCAL I NVARIANCES

文献摘要

相似文献