mlsanityData to 14 Oct 2026

Toward Learning Human-aligned Cross-domain Robust Models by Countering Misaligned Features

arXiv 2111.03740 ↗First posted 5 Nov 2021cs.LGcited by 14

Not on the map. Why?

Abstract (arXiv)

Machine learning has demonstrated remarkable prediction accuracy over i.i.d data, but the accuracy often drops when tested with data from another distribution. In this paper, we aim to offer another view of this problem in a perspective assuming the reason behind this accuracy drop is the reliance of models on the features that are not aligned well with how a data annotator considers similar across these two datasets. We refer to these features as misaligned features. We extend the conventional generalization error bound to a new one for this setup with the knowledge of how the misaligned features are associated with the label. Our analysis offers a set of techniques for this problem, and these techniques are naturally linked to many previous methods in robust machine learning literature. We also compared the empirical strength of these methods demonstrated the performance when these previous techniques are combined, with an implementation available at https://github.com/OoDBag/WR

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 44 of its 68 references that are in the map. First posted on arXiv 5 Nov 2021.

Builds on

44 of 68 references.

References · 44

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  4. 2013Intriguing properties of neural networkscited by 5,155
  5. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  6. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  7. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  8. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,628
  9. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
  10. 2019Natural Adversarial Examplescited by 1,408
  11. 2019Learning Robust Global Representations by Penalizing Local Predictive Powercited by 1,013
  12. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 953
32 more
  1. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 901
  2. 2020Self-Challenging Improves Cross-Domain Generalizationcited by 426
  3. 2019Bridging Theory and Algorithm for Domain Adaptationcited by 332
  4. 2017Unified Deep Supervised Domain Adaptation and Generalizationcited by 323
  5. 2019High Frequency Component Helps Explain the Generalization of Convolutional Neural Networkscited by 310
  6. 2017Right for the Right Reasons: Training Differentiable Models by Constraining their Explanationscited by 294
  7. 2019Mitigating Gender Bias in Natural Language Processing: Literature Reviewcited by 287
  8. 2019Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biasescited by 280
  9. 2018Generalizing Across Domains via Cross-Gradient Trainingcited by 268
  10. 2020Fairness without Demographics through Adversarially Reweighted Learningcited by 251
  11. 2019Learning De-biased Representations with Biased Representationscited by 199
  12. 2019RUBi: Reducing Unimodal Biases in Visual Question Answeringcited by 185
  13. 2016Statistics of Robust Optimization: A Generalized Empirical Likelihood Approachcited by 183
  14. 2016Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptationcited by 163
  15. 2020Making deep neural networks right for the right scientific reasons by interacting with their explanationscited by 142
  16. 2020Learning from Failure: Training Debiased Classifier from Biased Classifiercited by 133
  17. 2019End-to-End Bias Mitigation by Modelling Biases in Corporacited by 128
  18. 2019Learning Robust Representations by Projecting Superficial Statistics Outcited by 123
  19. 2019Unlearn Dataset Bias in Natural Language Inference by Fitting the Residualcited by 119
  20. 2019Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNscited by 90
  21. 2017Measuring the tendency of CNNs to Learn Surface Statistical Regularitiescited by 87
  22. 2016Beyond Sharing Weights for Deep Domain Adaptationcited by 80
  23. 2016Towards an integration of deep learning and neurosciencecited by 78
  24. 2020Estimating Generalization under Distribution Shifts via Domain-Invariant Representationscited by 56
  25. 2020FIND: Human-in-the-Loop Debugging Deep Text Classifierscited by 40
  26. 2019Learning From Brains How to Regularize Machinescited by 38
  27. 2018What made you do this? Understanding black-box decisions with sufficient input subsetscited by 31
  28. 2015A New PAC-Bayesian Perspective on Domain Adaptationcited by 29
  29. 2017Biologically inspired protection of deep networks from adversarial attackscited by 23
  30. 2021Removing Undesirable Feature Contributions Using Out-of-Distribution Datacited by 23
  31. 2020The Struggles of Feature-Based Explanations: Shapley Values vs. Minimal Sufficient Subsetscited by 14
  32. 2018Hallucinating Agnostic Images to Generalize Across Domainscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 2; 2022: 4; 2023: 4; 2024: 3; 2025: 0; 2026: 1.

14 papers.

Sort:

By area:

  1. 2021Measure and Improve Robustness in NLP Models: A Surveycited by 107
  2. 2021Towards Principled Disentanglement for Domain Generalizationcited by 91
  3. 2023ChatGPT to Replace Crowdsourcing of Paraphrases for Intent Classification: Higher Diversity and Comparable Model Robustnesscited by 41
  4. 2023A Sentence Speaks a Thousand Images: Domain Generalization through Distilling CLIP with Language Guidancecited by 40
  5. 2023Towards Trustworthy and Aligned Machine Learning: A Data-centric Survey with Causality Perspectivescited by 30
  6. 2022Exploring Transformer Backbones for Heterogeneous Treatment Effect Estimationcited by 25
  7. 2022The Two Dimensions of Worst-case Training and the Integrated Effect for Out-of-domain Generalizationcited by 21
  8. 2024Effects of diversity incentives on sample diversity and downstream model performance in LLM-based text augmentationcited by 17
  9. 2023Foundation Model-oriented Robustness: Robust Image Model Evaluation with Pretrained Modelscited by 6
  10. 2024MADOD: Generalizing OOD Detection to Unseen Domains via G-Invariance Meta-Learningcited by 5
  11. 2026Mind the Discriminability Trap in Source-Free Cross-domain Few-shot Learningcited by 4
  12. 2022Robustar: Interactive Toolbox Supporting Precise Data Annotation for Robust Vision Learningcited by 0
2 more
  1. 2022Expeditious Saliency-guided Mix-up through Random Gradient Thresholdingcited by 0
  2. 2024Quantitative Evaluation of the Saliency Map for Alzheimer's Disease Classifier with Anatomical Segmentationcited by 0