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 2108.02883stat.MLcs.LG
收藏

即使在没有噪声的情况下,插值也会损害鲁棒泛化能力

Interpolation can hurt robust generalization even when there is no noise

ETH Zurich(苏黎世联邦理工学院);Rice University(莱斯大学);Technical University of Munich(慕尼黑工业大学)

Konstantin Donhauser, Alexandru Ţifrea, Michael Aerni, Reinhard Heckel, Fanny Yang

摘要:

最近的许多工作表明,overparameterization隐式减少方差的最小范数插值器和最大利润分类。这些发现表明,脊正则化在高维中具有消失的好处。我们通过证明即使在没有噪声的情况下,通过岭正则化避免插值也可以显着提高泛化能力来挑战这种说法。我们证明了这一现象的鲁棒风险的线性回归和分类,从而提供了第一个理论结果的鲁棒过拟合。

英文摘要:

Numerous recent works show that overparameterization implicitly reduces variance for min-norm interpolators and max-margin classifiers. These findings suggest that ridge regularization has vanishing benefits in high dimensions. We challenge this narrative by showing that, even in the absence of noise, avoiding interpolation through ridge regularization can significantly improve generalization. We prove this phenomenon for the robust risk of both linear regression and classification and hence provide the first theoretical result on robust overfitting.

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。

↑