arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.01008cs.LGcs.CV
收藏

通过重写预测规则来编辑分类器

Editing a classifier by rewriting its prediction rules

MIT(麻省理工学院)

Shibani Santurkar, Dimitris Tsipras, Mahalaxmi Elango, David Bau, Antonio Torralba, Aleksander Madry

摘要:

我们提出了一种方法来修改的行为的分类器直接重写其预测规则。我们的方法几乎不需要额外的数据汇集,可以应用于各种设置,包括使模型适应新的环境,并对其进行修改以忽略虚假特征。我们的代码可在https://github.com/MadryLab/EditingClassifiers上获得。

英文摘要:

We present a methodology for modifying the behavior of a classifier by directly rewriting its prediction rules. Our approach requires virtually no additional data collection and can be applied to a variety of settings, including adapting a model to new environments, and modifying it to ignore spurious features. Our code is available at https://github.com/MadryLab/EditingClassifiers .

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。

↑