Evading the Simplicity Bias: Training a Diverse Set of Models Discovers Solutions with Superior OOD Generalization
Author:
Affiliation:
1. Idiap Research Institute
2. University of Adelaide,Australian Institute for Machine Learning
Publisher
IEEE
Link
http://xplorestaging.ieee.org/ielx7/9878378/9878366/09878716.pdf?arnumber=9878716
Reference88 articles.
1. Axiomatic attribution for deep networks;sundararajan;Proc Int Conf Mach Learn,0
2. Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training
3. Unbiased look at dataset bias;antonio;CVPR,2011
Cited by 15 articles. 订阅此论文施引文献 订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
1. Fake news detection: Taxonomy and comparative study;Information Fusion;2024-03
2. DoubleAUG: Single-domain Generalized Object Detector in Urban via Color Perturbation and Dual-style Memory;ACM Transactions on Multimedia Computing, Communications, and Applications;2024-01-11
3. Using Early Readouts to Mediate Featural Bias in Distillation;2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV);2024-01-03
4. Model Guidance via Explanations Turns Image Classifiers into Segmentation Models;Communications in Computer and Information Science;2024
5. Attribute-Aware Deep Hashing With Self-Consistency for Large-Scale Fine-Grained Image Retrieval;IEEE Transactions on Pattern Analysis and Machine Intelligence;2023-11-01
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3