Overconfidence in the Face of Ambiguity with Adversarial Data

Margaret Li, Julian Michael
{"title":"Overconfidence in the Face of Ambiguity with Adversarial Data","authors":"Margaret Li, Julian Michael","doi":"10.18653/v1/2022.dadc-1.4","DOIUrl":null,"url":null,"abstract":"Adversarial data collection has shown promise as a method for building models which are more robust to the spurious correlations that generally appear in naturalistic data. However, adversarially-collected data may itself be subject to biases, particularly with regard to ambiguous or arguable labeling judgments. Searching for examples where an annotator disagrees with a model might over-sample ambiguous inputs, and filtering the results for high inter-annotator agreement may under-sample them. In either case, training a model on such data may produce predictable and unwanted biases. In this work, we investigate whether models trained on adversarially-collected data are miscalibrated with respect to the ambiguity of their inputs. Using Natural Language Inference models as a testbed, we find no clear difference in accuracy between naturalistically and adversarially trained models, but our model trained only on adversarially-sourced data is considerably more overconfident of its predictions and demonstrates worse calibration, especially on ambiguous inputs. This effect is mitigated, however, when naturalistic and adversarial training data are combined.","PeriodicalId":214275,"journal":{"name":"Proceedings of the First Workshop on Dynamic Adversarial Data Collection","volume":"90 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the First Workshop on Dynamic Adversarial Data Collection","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.18653/v1/2022.dadc-1.4","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Adversarial data collection has shown promise as a method for building models which are more robust to the spurious correlations that generally appear in naturalistic data. However, adversarially-collected data may itself be subject to biases, particularly with regard to ambiguous or arguable labeling judgments. Searching for examples where an annotator disagrees with a model might over-sample ambiguous inputs, and filtering the results for high inter-annotator agreement may under-sample them. In either case, training a model on such data may produce predictable and unwanted biases. In this work, we investigate whether models trained on adversarially-collected data are miscalibrated with respect to the ambiguity of their inputs. Using Natural Language Inference models as a testbed, we find no clear difference in accuracy between naturalistically and adversarially trained models, but our model trained only on adversarially-sourced data is considerably more overconfident of its predictions and demonstrates worse calibration, especially on ambiguous inputs. This effect is mitigated, however, when naturalistic and adversarial training data are combined.
面对模棱两可的敌对数据时的过度自信
对抗性数据收集作为一种建立模型的方法已经显示出了希望,这种模型对通常出现在自然数据中的虚假相关性更为稳健。然而,对抗性收集的数据本身可能会受到偏见的影响,特别是在模棱两可或有争议的标签判断方面。搜索注释者不同意模型的示例可能会对模糊输入进行过度采样,而过滤注释者之间高度一致的结果可能会对它们进行欠采样。在任何一种情况下,在这些数据上训练模型都可能产生可预见的和不必要的偏差。在这项工作中,我们研究了在对抗性收集的数据上训练的模型是否因其输入的模糊性而被错误校准。使用自然语言推理模型作为测试平台,我们发现自然和对抗训练模型之间的准确性没有明显差异,但是我们的模型仅在对抗来源的数据上训练,对其预测过于自信,并且显示出更差的校准,特别是在模糊输入上。然而,当自然训练数据和对抗训练数据结合在一起时,这种影响就会减轻。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书