mlsanityData to 14 Oct 2026

TTAPS: Test-Time Adaption by Aligning Prototypes using Self-Supervision

arXiv 2205.08731 ↗First posted 18 May 2022cs.LG cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Nowadays, deep neural networks outperform humans in many tasks. However, if the input distribution drifts away from the one used in training, their performance drops significantly. Recently published research has shown that adapting the model parameters to the test sample can mitigate this performance degradation. In this paper, we therefore propose a novel modification of the self-supervised training algorithm SwAV that adds the ability to adapt to single test samples. Using the provided prototypes of SwAV and our derived test-time loss, we align the representation of unseen test samples with the self-supervised learned prototypes. We show the success of our method on the common benchmark dataset CIFAR10-C.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 40 of its 47 references that are in the map. First posted on arXiv 18 May 2022.

Builds on

40 of 47 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  4. 2013Intriguing properties of neural networkscited by 5,155
  5. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  6. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  7. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  8. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  9. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  10. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
  11. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  12. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
28 more
  1. 2016Matching Networks for One Shot Learningcited by 2,255
  2. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,850
  3. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,845
  4. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,591
  5. 2018Group Normalizationcited by 1,524
  6. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  7. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,143
  8. 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 808
  9. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 742
  10. 2019Data-Efficient Image Recognition with Contrastive Predictive Codingcited by 673
  11. 2018Deep Visual Domain Adaptation: A Surveycited by 635
  12. 2017Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methodscited by 546
  13. 2021Towards Out-Of-Distribution Generalization: A Surveycited by 436
  14. 2020Improving robustness against common corruptions by covariate shift adaptationcited by 431
  15. 2018A Survey on Deep Transfer Learningcited by 419
  16. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 377
  17. 2018A Survey of Unsupervised Deep Domain Adaptationcited by 361
  18. 2021Fixing Data Augmentation to Improve Adversarial Robustnesscited by 204
  19. 2018Why do deep convolutional networks generalize so poorly to small image transformations?cited by 198
  20. 2020Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuningcited by 153
  21. 2020A Review of Single-Source Deep Unsupervised Visual Domain Adaptationcited by 96
  22. 2015Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networkscited by 95
  23. 2021MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaptioncited by 65
  24. 2020Adversarial Style Mining for One-Shot Unsupervised Domain Adaptationcited by 64
  25. 2020A Framework For Contrastive Self-Supervised Learning And Designing A New Approachcited by 52
  26. 2018One-Shot Unsupervised Cross Domain Translationcited by 35
  27. 2020Learn2Perturb: an End-to-end Feature Perturbation Learning to Improve Adversarial Robustnesscited by 27
  28. 2020Improving out-of-distribution generalization via multi-task self-supervised pretrainingcited by 26

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 2; 2024: 5.

8 papers.

  1. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 359
  2. 2022Robust Mean Teacher for Continual and Gradual Test-Time Adaptationcited by 104
  3. 2024Protected Test-Time Adaptation via Online Entropy Matching: A Betting Approachcited by 23
  4. 2023What, How, and When Should Object Detectors Update in Continually Changing Test Domains?cited by 15
  5. 2024Test-time Adaptation Meets Image Enhancement: Improving Accuracy via Uncertainty-aware Logit Switchingcited by 2
  6. 2024Test-time Training for Hyperspectral Image Super-resolutioncited by 2
  7. 2024Less is More: Pseudo-Label Filtering for Continual Test-Time Adaptationcited by 1
  8. 2024Mitigating the Bias in the Model for Continual Test-Time Adaptationcited by 0