mlsanityData to 14 Oct 2026

Unraveling Meta-Learning: Understanding Feature Representations for Few-Shot Tasks

arXiv 2002.06753 ↗First posted 17 Feb 2020cs.LG cs.CV stat.MLcited by 47

Not on the map. Why?

Abstract (arXiv)

Meta-learning algorithms produce feature extractors which achieve state-of-the-art performance on few-shot classification. While the literature is rich with meta-learning methods, little is known about why the resulting feature extractors perform so well. We develop a better understanding of the underlying mechanics of meta-learning and the difference between models trained using meta-learning and models which are trained classically. In doing so, we introduce and verify several hypotheses for why meta-learned models perform better. Furthermore, we develop a regularizer which boosts the performance of standard training routines for few-shot classification. In many cases, our routine outperforms meta-learning while simultaneously running an order of magnitude faster.

Key facts. Cited by 47 papers in the map (data through 14 Oct 2026). First posted on arXiv 17 Feb 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 15; 2022: 15; 2023: 6; 2024: 5; 2025: 2; 2026: 4.

47 papers.

Sort:

By area:

  1. 2022Task Residual for Tuning Vision-Language Modelscited by 123
  2. 2021Simpler is Better: Few-shot Semantic Segmentation with Classifier Weight Transformercited by 116
  3. 2024What makes unlearning hard and what to do about itcited by 89
  4. 2021Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective Adaptationcited by 50
  5. 2021Meta-Learning with Task-Adaptive Loss Function for Few-Shot Learningcited by 49
  6. 2021Shallow Bayesian Meta Learning for Real-World Few-Shot Recognitioncited by 48
  7. 2021Orthogonal Projection Losscited by 44
  8. 2023Reverse Engineering Self-Supervised Learningcited by 33
  9. 2021Binocular Mutual Learning for Improving Few-shot Classificationcited by 26
  10. 2022Deconfounded Representation Similarity for Comparison of Neural Networkscited by 19
  11. 2024Imperative Learning: A Self-supervised Neuro-Symbolic Learning Framework for Robot Autonomycited by 16
  12. 2021On sensitivity of meta-learning to support datacited by 15
35 more
  1. 2024Uncertainty Herding: One Active Learning Method for All Label Budgetscited by 15
  2. 2021A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learningcited by 12
  3. 2021Meta-learning with negative learning ratescited by 11
  4. 2022Scalable PAC-Bayesian Meta-Learning via the PAC-Optimal Hyper-Posterior: From Theory to Practicecited by 11
  5. 2022The Curse of Low Task Diversity: On the Failure of Transfer Learning to Outperform MAML and Their Empirical Equivalencecited by 10
  6. 2023Open-Set Likelihood Maximization for Few-Shot Learningcited by 10
  7. 2021Bridging Few-Shot Learning and Adaptation: New Challenges of Support-Query Shiftcited by 9
  8. 2021MAML is a Noisy Contrastive Learner in Classificationcited by 9
  9. 2022FS-BAN: Born-Again Networks for Domain Generalization Few-Shot Classificationcited by 9
  10. 2022Generalization Bounds for Few-Shot Transfer Learning with Pretrained Classifierscited by 7
  11. 2026Contrastive Representation Shaping for LLM Unlearningcited by 7
  12. 2021A Channel Coding Benchmark for Meta-Learningcited by 6
  13. 2023Is Pre-training Truly Better Than Meta-Learning?cited by 6
  14. 2024BTBR: A Bayesian-Theory-Driven Probabilistic-Fuzzy Framework for Implicit Bias Removal in Large Language Modelscited by 4
  15. 2025Is Meta-Learning Out? Rethinking Unsupervised Few-Shot Classification with Limited Entropycited by 4
  16. 2021Towards Enabling Meta-Learning from Target Modelscited by 3
  17. 2026Directional Neural Collapse Explains Few-Shot Transfer in Self-Supervised Learningcited by 3
  18. 2022Meta-free few-shot learning via representation learning with weight averagingcited by 2
  19. 2022Model-Agnostic Few-Shot Open-Set Recognitioncited by 2
  20. 2022Improving Meta-Learning Generalization with Activation-Based Early-Stoppingcited by 2
  21. 2022Robust Meta-Representation Learning via Global Label Inference and Classificationcited by 2
  22. 2023Meta-Auxiliary Network for 3D GAN Inversioncited by 2
  23. 2021Meta-learning representations for clustering with infinite Gaussian mixture modelscited by 1
  24. 2021Meta-learning an Intermediate Representation for Few-shot Block-wise Prediction of Landslide Susceptibilitycited by 1
  25. 2022Negative Inner-Loop Learning Rates Learn Universal Featurescited by 1
  26. 2023Exploring Active Learning in Meta-Learning: Enhancing Context Set Labelingcited by 1
  27. 2022AutoProtoNet: Interpretability for Prototypical Networkscited by 0
  28. 2022Diagnosing and Remedying Shot Sensitivity with Cosine Few-Shot Learnerscited by 0
  29. 2022On the Subspace Structure of Gradient-Based Meta-Learningcited by 0
  30. 2022Not All Instances Contribute Equally: Instance-adaptive Class Representation Learning for Few-Shot Visual Recognitioncited by 0
  31. 2023Improving Generalization in Meta-Learning via Meta-Gradient Augmentationcited by 0
  32. 2024Unsupervised Meta-Learning via Dynamic Head and Heterogeneous Task Construction for Few-Shot Classificationcited by 0
  33. 2025Neural Coherence : Find higher performance to out-of-distribution tasks from few samplescited by 0
  34. 2026Information Theoretic Perspective on Representation Learningcited by 0
  35. 2026Which Tasks Survive Self-Supervised Learning?cited by 0