mlsanityData to 14 Oct 2026

Meta-Learning Requires Meta-Augmentation

arXiv 2007.05549 ↗First posted 10 Jul 2020cs.LG stat.MLcited by 57

Not on the map. Why?

Abstract (arXiv)

Meta-learning algorithms aim to learn two components: a model that predicts targets for a task, and a base learner that quickly updates that model when given examples from a new task. This additional level of learning can be powerful, but it also creates another potential source for overfitting, since we can now overfit in either the model or the base learner. We describe both of these forms of metalearning overfitting, and demonstrate that they appear experimentally in common meta-learning benchmarks. We then use an information-theoretic framework to discuss meta-augmentation, a way to add randomness that discourages the base learner and model from learning trivial solutions that do not generalize to new tasks. We demonstrate that meta-augmentation produces large complementary benefits to recently proposed meta-regularization techniques.

Key facts. Cited by 57 papers in the map (data through 14 Oct 2026). First posted on arXiv 10 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 12; 2022: 20; 2023: 9; 2024: 12; 2025: 4.

57 papers.

Sort:

By area:

  1. 2023Symbol tuning improves in-context learning in language modelscited by 75
  2. 2022C-Mixup: Improving Generalization in Regressioncited by 57
  3. 2021Meta-Learning with Fewer Tasks through Task Interpolationcited by 45
  4. 2021Same State, Different Task: Continual Reinforcement Learning without Interferencecited by 43
  5. 2021Augmented World Models Facilitate Zero-Shot Dynamics Generalization From a Single Offline Environmentcited by 42
  6. 2022Fixing Model Bugs with Natural Language Patchescited by 33
  7. 2021Meta-learning with an Adaptive Task Schedulercited by 27
  8. 2022Meta Learning for Natural Language Processing: A Surveycited by 26
  9. 2021How to Train Your MAML to Excel in Few-Shot Classificationcited by 22
  10. 2024Position: Leverage Foundational Models for Black-Box Optimizationcited by 21
  11. 2021Few-Shot Learning with Class Imbalancecited by 18
  12. 2022What Matters For Meta-Learning Vision Regression Tasks?cited by 18
38 more
  1. 2022Learning Deep Time-index Models for Time Series Forecastingcited by 18
  2. 2022Neural Processes with Stochastic Attention: Paying more attention to the context datasetcited by 16
  3. 2022Model-Based Offline Meta-Reinforcement Learning with Regularizationcited by 15
  4. 2024Robust Fast Adaptation from Adversarially Explicit Task Distribution Generationcited by 13
  5. 2025Evolution and The Knightian Blindspot of Machine Learningcited by 11
  6. 2021MAML is a Noisy Contrastive Learner in Classificationcited by 9
  7. 2022Improving Meta-learning for Low-resource Text Classification and Generation via Memory Imitationcited by 8
  8. 2023Hacking Task Confounder in Meta-Learningcited by 8
  9. 2021Multimodal Meta-Learning for Time Series Regressioncited by 7
  10. 2022Set-based Meta-Interpolation for Few-Task Meta-Learningcited by 7
  11. 2024AdaAugment: A Tuning-Free and Adaptive Approach to Enhance Data Augmentationcited by 7
  12. 2024Meta-GPS++: Enhancing Graph Meta-Learning with Contrastive Learning and Self-Trainingcited by 7
  13. 2022Diverse Preference Augmentation with Multiple Domains for Cold-start Recommendationscited by 6
  14. 2022Robust Meta-learning with Sampling Noise and Label Noise via Eigen-Reptilecited by 6
  15. 2025Instance Generation for Meta-Black-Box Optimization through Latent Space Reverse Engineeringcited by 6
  16. 2022A Meta-GNN approach to personalized seizure detection and classificationcited by 5
  17. 2025Dual-level Mixup for Graph Few-shot Learning with Fewer Taskscited by 5
  18. 2023DAC-MR: Data Augmentation Consistency Based Meta-Regularization for Meta-Learningcited by 4
  19. 2023Task-Distributionally Robust Data-Free Meta-Learningcited by 4
  20. 2021Semi-Supervised Few-Shot Intent Classification and Slot Fillingcited by 3
  21. 2021Ortho-Shot: Low Displacement Rank Regularization with Data Augmentation for Few-Shot Learningcited by 3
  22. 2023Meta-Optimization for Higher Model Generalizability in Single-Image Depth Predictioncited by 3
  23. 2024Unsupervised Meta-Learning via In-Context Learningcited by 3
  24. 2025Few-Shot Learning in Video and 3D Object Detection: A Surveycited by 2
  25. 2021Meta-Regularization by Enforcing Mutual-Exclusivenesscited by 1
  26. 2021On the inductive biases of deep domain adaptationcited by 1
  27. 2022Towards Better Meta-Initialization with Task Augmentation for Kindergarten-aged Speech Recognitioncited by 1
  28. 2022Naive Few-Shot Learning: Uncovering the fluid intelligence of machinescited by 1
  29. 2024Few-Shot Load Forecasting Under Data Scarcity in Smart Grids: A Meta-Learning Approachcited by 1
  30. 2024Rethinking Meta-Learning from a Learning Lenscited by 1
  31. 2022Zero-Episode Few-Shot Contrastive Predictive Coding: Solving intelligence tests without prior trainingcited by 0
  32. 2022Persian Natural Language Inference: A Meta-learning approachcited by 0
  33. 2022Learning to generate imaginary tasks for improving generalization in meta-learningcited by 0
  34. 2022A new benchmark for group distribution shifts in hand grasp regression for object manipulation. Can meta-learning raise the bar?cited by 0
  35. 2022Wormhole MAML: Meta-Learning in Glued Parameter Spacecited by 0
  36. 2022Smooth Mathematical Function from Compact Neural Networkscited by 0
  37. 2023AwesomeMeta+: A Mixed-Prototyping Meta-Learning System Supporting AI Application Design Anywherecited by 0
  38. 2023Improving Generalization in Meta-Learning via Meta-Gradient Augmentationcited by 0

50 most cited of 57 All 57 (JSON)