arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.04237cs.CVcs.AIcs.LG
收藏

通过解释学习鲁棒的卷积神经网络以聚焦相关特征

Learning Robust Convolutional Neural Networks with Relevant Feature Focusing via Explanations

NTT Corporation(日本电信电话株式会社)

Kazuki Adachi, Shin'ya Yamaguchi

AI总结:

本文提出ReFF方法,通过解释输出正则化CNN,使其聚焦于任务相关特征,从而提升模型在分布偏移下的鲁棒性和测试精度。

摘要:

现有的基于卷积神经网络(CNN)的图像识别技术基本上假设训练集和测试集是从独立同分布(i.i.d)分布中采样的。然而,在现实世界中,这种假设很容易被打破,因为当输入图像中物体与背景之间的共现关系发生变化时,会出现分布偏移。在这种类型的分布偏移下,CNNs会学习聚焦于不相关的特征,如训练数据中的背景,从而在测试数据上降低准确性。为了解决这个问题,我们提出了相关特征聚焦(ReFF)。ReFF通过解释输出(例如Grad-CAM)检测任务相关的特征,并通过解释输出对CNN进行正则化。由于ReFF由事后解释模块组成,因此可以轻松地应用于现成的CNNs。此外,ReFF在测试时不需要额外的推理成本,因为它仅在训练时用于正则化。我们证明了通过ReFF训练的CNNs会聚焦于与目标任务相关的特征,并且ReFF提高了测试时的准确性。

英文摘要:

Existing image recognition techniques based on convolutional neural networks (CNNs) basically assume that the training and test datasets are sampled from i.i.d distributions. However, this assumption is easily broken in the real world because of the distribution shift that occurs when the co-occurrence relations between objects and backgrounds in input images change. Under this type of distribution shift, CNNs learn to focus on features that are not task-relevant, such as backgrounds from the training data, and degrade their accuracy on the test data. To tackle this problem, we propose relevant feature focusing (ReFF). ReFF detects task-relevant features and regularizes CNNs via explanation outputs (e.g., Grad-CAM). Since ReFF is composed of post-hoc explanation modules, it can be easily applied to off-the-shelf CNNs. Furthermore, ReFF requires no additional inference cost at test time because it is only used for regularization while training. We demonstrate that CNNs trained with ReFF focus on features relevant to the target task and that ReFF improves the test-time accuracy.

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。

↑