分布稳健模型的元分析
A Meta-Analysis of Distributionally-Robust Models
摘要:
在大规模数据集(如ImageNet)上训练的最先进的图像分类器已被证明容易受到一系列有意和偶然的分布偏移的影响。另一方面,最近出现了几个具有良好的分布外(OOD)鲁棒性的分类器,在目标任务上实现了高精度,同时在具有挑战性的基准测试中保持了分布内的精度。我们对广泛的公开发布的模型进行了荟萃分析,其中大部分是在过去12个月内发布的。通过这种元分析,我们根据经验确定了所有表现最好的OOD鲁棒模型的四个主要共性,所有这些都说明了视觉语言预训练的巨大潜力。
英文摘要:
State-of-the-art image classifiers trained on massive datasets (such as ImageNet) have been shown to be vulnerable to a range of both intentional and incidental distribution shifts. On the other hand, several recent classifiers with favorable out-of-distribution (OOD) robustness properties have emerged, achieving high accuracy on their target tasks while maintaining their in-distribution accuracy on challenging benchmarks. We present a meta-analysis on a wide range of publicly released models, most of which have been published over the last twelve months. Through this meta-analysis, we empirically identify four main commonalities for all the best-performing OOD-robust models, all of which illuminate the considerable promise of vision-language pre-training.
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。