arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.04191cs.CVcs.AIcs.LG
收藏

SI-Score:一个图像数据集,用于对对象位置、旋转和大小的鲁棒性进行细粒度分析

SI-Score: An image dataset for fine-grained analysis of robustness to object location, rotation and size

Google Research, Brain Team(谷歌研究院,大脑团队)

Jessica Yung, Rob Romijnders, Alexander Kolesnikov, Lucas Beyer, Josip Djolonga, Neil Houlsby, Sylvain Gelly, Mario Lucic, Xiaohua Zhai

摘要:

在部署机器学习模型之前,评估其鲁棒性至关重要。在用于图像理解的深度神经网络的上下文中,改变对象位置、旋转和大小可能以非平凡的方式影响预测。在这项工作中,我们执行一个细粒度的分析,这些因素的变化使用SI分数,一个合成数据集的鲁棒性。特别是,我们研究了ResNets,视觉Transformer和CLIP,并确定了它们之间有趣的定性差异。

英文摘要:

Before deploying machine learning models it is critical to assess their robustness. In the context of deep neural networks for image understanding, changing the object location, rotation and size may affect the predictions in non-trivial ways. In this work we perform a fine-grained analysis of robustness with respect to these factors of variation using SI-Score, a synthetic dataset. In particular, we investigate ResNets, Vision Transformers and CLIP, and identify interesting qualitative differences between these.

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。

↑