mlsanityData to 7 Oct 2026

ReFine: Re-randomization before Fine-tuning for Cross-domain Few-shot Learning

arXiv 2205.05282 ↗First posted 11 May 2022cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Cross-domain few-shot learning (CD-FSL), where there are few target samples under extreme differences between source and target domains, has recently attracted huge attention. Recent studies on CD-FSL generally focus on transfer learning based approaches, where a neural network is pre-trained on popular labeled source domain datasets and then transferred to target domain data. Although the labeled datasets may provide suitable initial parameters for the target data, the domain difference between the source and target might hinder fine-tuning on the target domain. This paper proposes a simple yet powerful method that re-randomizes the parameters fitted on the source domain before adapting to the target data. The re-randomization resets source-specific parameters of the source pre-trained model and thus facilitates fine-tuning on the target domain, improving few-shot performance.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 35 of its 43 references that are in the map. First posted on arXiv 11 May 2022.

Builds on

35 of 43 references.

References · 35

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  2. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  3. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  4. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,250
  5. 2016Matching Networks for One Shot Learningcited by 2,246
  6. 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,488
  7. 2014How transferable are features in deep neural networks?cited by 1,216
  8. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  9. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  10. 2013Exact solutions to the nonlinear dynamics of learning in deep linear neural networkscited by 748
  11. 2019A Closer Look at Few-shot Classificationcited by 679
  12. 2019Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)cited by 657
23 more
  1. 2016Learning to learn by gradient descent by gradient descentcited by 616
  2. 2019Generalizing from a Few Examples: A Survey on Few-Shot Learningcited by 462
  3. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  4. 2020Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?cited by 435
  5. 2019Transfusion: Understanding Transfer Learning for Medical Imagingcited by 366
  6. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  7. 2019A Baseline for Few-Shot Image Classificationcited by 298
  8. 2018Probabilistic Model-Agnostic Meta-Learningcited by 254
  9. 2020Revisiting Few-sample BERT Fine-tuningcited by 237
  10. 2016Using Deep Learning for Image-Based Plant Disease Detectioncited by 216
  11. 2020Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformationcited by 183
  12. 2019Meta-Learning with Warped Gradient Descentcited by 98
  13. 2019Are All Layers Created Equal?cited by 94
  14. 2020Self-training for Few-shot Transfer Across Extreme Task Differencescited by 75
  15. 2020What Do Neural Networks Learn When Trained With Random Labels?cited by 58
  16. 2021Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencodercited by 51
  17. 2021Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Datacited by 49
  18. 2020Investigating Transferability in Pretrained Language Modelscited by 28
  19. 2020RIFLE: Backpropagation in Depth for Deep Transfer Learning through Re-Initializing the Fully-connected LayErcited by 27
  20. 2021The Impact of Reinitialization on Generalization in Convolutional Neural Networkscited by 20
  21. 2021Multi-level Metric Learning for Few-shot Image Recognitioncited by 8
  22. 2019HIDRA: Head Initialization across Dynamic targets for Robust Architecturescited by 3
  23. 2019Backbone Can Not be Trained at Once: Rolling Back to Pre-trained Network for Person Re-Identificationcited by 2

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 3; 2025: 1.

6 papers.

  1. 2022A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunitiescited by 162
  2. 2024Enhancing Information Maximization with Distance-Aware Contrastive Learning for Source-Free Cross-Domain Few-Shot Learningcited by 13
  3. 2024Towards Few-Shot Learning in the Open World: A Review and Beyondcited by 5
  4. 2024Step-wise Distribution Alignment Guided Style Prompt Tuning for Source-free Cross-domain Few-shot Learningcited by 5
  5. 2023A Survey of Deep Visual Cross-Domain Few-Shot Learningcited by 3
  6. 2025FreqGRL: Suppressing Low-Frequency Bias and Mining High-Frequency Knowledge for Cross-Domain Few-Shot Learningcited by 0