mlsanityData to 13 Oct 2026

Out-of-Distribution Generalization Analysis via Influence Function

arXiv 2101.08521 ↗First posted 21 Jan 2021cs.LG stat.MLcited by 11

Not on the map. Why?

Abstract (arXiv)

The mismatch between training and target data is one major challenge for current machine learning systems. When training data is collected from multiple domains and the target domains include all training domains and other new domains, we are facing an Out-of-Distribution (OOD) generalization problem that aims to find a model with the best OOD accuracy. One of the definitions of OOD accuracy is worst-domain accuracy. In general, the set of target domains is unknown, and the worst over target domains may be unseen when the number of observed domains is limited. In this paper, we show that the worst accuracy over the observed domains may dramatically fail to identify the OOD accuracy. To this end, we introduce Influence Function, a classical tool from robust statistics, into the OOD generalization problem and suggest the variance of influence function to monitor the stability of a model on training domains. We show that the accuracy on test domains and the proposed index together can help us discern whether OOD algorithms are needed and whether a model achieves good OOD generalization.

Key facts. Cited by 11 papers in the map (data through 13 Oct 2026). Builds on 24 of its 47 references that are in the map. First posted on arXiv 21 Jan 2021.

Builds on

24 of 47 references.

References · 24

  1. 2019Invariant Risk Minimizationcited by 1,842
  2. 2017Understanding Black-box Predictions via Influence Functionscited by 1,729
  3. 2020In Search of Lost Domain Generalizationcited by 1,048
  4. 2017Deeper, Broader and Artier Domain Generalizationcited by 984
  5. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 738
  6. 2018Recognition in Terra Incognitacited by 656
  7. 2015Causal inference using invariant prediction: identification and confidence intervalscited by 534
  8. 2019Adversarial Domain Adaptation with Domain Mixupcited by 250
  9. 2020Influence Functions in Deep Learning Are Fragilecited by 249
  10. 2019Causality matters in medical imagingcited by 193
  11. 2020Invariant Risk Minimization Gamescited by 188
  12. 2020Improve Unsupervised Domain Adaptation with Mixup Trainingcited by 185
12 more
  1. 2019A Meta-Transfer Objective for Learning to Disentangle Causal Mechanismscited by 158
  2. 2019On the Accuracy of Influence Functions for Measuring Group Effectscited by 146
  3. 2017Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributionscited by 125
  4. 2018Invariance, Causality and Robustnesscited by 123
  5. 2020Heterogeneous Domain Generalization via Domain Mixupcited by 103
  6. 2019Detecting Adversarial Samples Using Influence Functions and Nearest Neighborscited by 72
  7. 2020Stable Prediction with Model Misspecification and Agnostic Distribution Shiftcited by 62
  8. 2020Influence Function based Data Poisoning Attacks to Top-N Recommender Systemscited by 62
  9. 2017Model-Powered Conditional Independence Testcited by 43
  10. 2019CCMI : Classifier based Conditional Mutual Information Estimationcited by 37
  11. 2020Risk Variance Penalizationcited by 21
  12. 2018Explaining Latent Factor Models for Recommendation with Influence Functionscited by 14

Built on by

Cited by year (2026 to 13 Oct)

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

11 papers.

  1. 2021Towards Out-Of-Distribution Generalization: A Surveycited by 432
  2. 2021Fishr: Invariant Gradient Variances for Out-of-Distribution Generalizationcited by 186
  3. 2021Towards a Theoretical Framework of Out-of-Distribution Generalizationcited by 94
  4. 2021Understanding Instance-based Interpretability of Variational Auto-Encoderscited by 21
  5. 2022A Review of Causality for Learning Algorithms in Medical Image Analysiscited by 18
  6. 2024Label Smoothing Improves Machine Unlearningcited by 9
  7. 2023When Graph Neural Network Meets Causality: Opportunities, Methodologies and An Outlookcited by 4
  8. 2023Evaluating the Impact of Local Differential Privacy on Utility Loss via Influence Functionscited by 1
  9. 2026Causal Transfer in Medical Image Analysiscited by 0
  10. 2026PLSP (Pre-hoc Liminal Space Profiling): OOD Prediction over Detection -- An Anticipatory Approach for Machine Learning Model Reliabilitycited by 0
  11. 2026Can Circuit Alignment Predict OOD Generalization?cited by 0