arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2202.07552cs.LG
收藏

变换不变性下的PAC学习理论

A Theory of PAC Learnability under Transformation Invariances

Toyota Technological Institute Chicago(芝加哥丰田技术研究院)

Han Shao, Omar Montasser, Avrim Blum

摘要:

变换不变性存在于许多现实问题中。例如,图像分类通常对旋转和颜色变换是不变的:旋转后的不同颜色的汽车仍然被识别为汽车。数据增强将转换后的数据添加到训练集中,并在增强后的数据上训练模型,这是一种将这些不变性构建到学习过程中的常用技术。然而,目前还不清楚如何数据增强执行理论上和最佳算法是在变换不变性的存在。在本文中,我们研究了PAC学习下的转换不变性在三个设置根据不同的可实现性水平:(i)一个假设适合增广的数据;(ii)一个假设只适合原始数据和转换后的数据躺在数据分布的支持;(iii)不可知的情况下。一个有趣的观察是,在设置(ii)和(iii)中,必须区分原始数据和变换数据以实现最佳精度,这意味着任何不区分原始数据和变换数据的算法(包括数据增强)都不是最佳的。此外,这种类型的算法甚至会“损害”准确性。在设置(i)中,尽管不需要区分两个数据集,但数据增强仍然不能最佳地执行。由于这样的差异,我们提出了两个组合的措施,其特征在于设置(i)和(ii)(iii)的最佳样本复杂度,并提供了最佳算法。

英文摘要:

Transformation invariances are present in many real-world problems. For example, image classification is usually invariant to rotation and color transformation: a rotated car in a different color is still identified as a car. Data augmentation, which adds the transformed data into the training set and trains a model on the augmented data, is one commonly used technique to build these invariances into the learning process. However, it is unclear how data augmentation performs theoretically and what the optimal algorithm is in presence of transformation invariances. In this paper, we study PAC learnability under transformation invariances in three settings according to different levels of realizability: (i) A hypothesis fits the augmented data; (ii) A hypothesis fits only the original data and the transformed data lying in the support of the data distribution; (iii) Agnostic case. One interesting observation is that distinguishing between the original data and the transformed data is necessary to achieve optimal accuracy in setting (ii) and (iii), which implies that any algorithm not differentiating between the original and transformed data (including data augmentation) is not optimal. Furthermore, this type of algorithms can even "harm" the accuracy. In setting (i), although it is unnecessary to distinguish between the two data sets, data augmentation still does not perform optimally. Due to such a difference, we propose two combinatorial measures characterizing the optimal sample complexity in setting (i) and (ii)(iii) and provide the optimal algorithms.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑