mlsanityData to 7 Oct 2026

Is Bayesian Model-Agnostic Meta Learning Better than Model-Agnostic Meta Learning, Provably?

arXiv 2203.03059 ↗First posted 6 Mar 2022cs.LG stat.MLcited by 13

Not on the map. Why?

Abstract (arXiv)

Meta learning aims at learning a model that can quickly adapt to unseen tasks. Widely used meta learning methods include model agnostic meta learning (MAML), implicit MAML, Bayesian MAML. Thanks to its ability of modeling uncertainty, Bayesian MAML often has advantageous empirical performance. However, the theoretical understanding of Bayesian MAML is still limited, especially on questions such as if and when Bayesian MAML has provably better performance than MAML. In this paper, we aim to provide theoretical justifications for Bayesian MAML's advantageous performance by comparing the meta test risks of MAML and Bayesian MAML. In the meta linear regression, under both the distribution agnostic and linear centroid cases, we have established that Bayesian MAML indeed has provably lower meta test risks than MAML. We verify our theoretical results through experiments.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 41 of its 61 references that are in the map. First posted on arXiv 6 Mar 2022.

Builds on

41 of 61 references.

References · 41

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  2. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  3. 2016Matching Networks for One Shot Learningcited by 2,246
  4. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  5. 2020Meta-Learning in Neural Networks: A Surveycited by 944
  6. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 762
  7. 2016Learning to learn by gradient descent by gradient descentcited by 616
  8. 2019Meta-Learning with Implicit Gradientscited by 523
  9. 2018Bilevel Programming for Hyperparameter Optimization and Meta-Learningcited by 403
  10. 2018Probabilistic Model-Agnostic Meta-Learningcited by 254
  11. 2018Meta-Learning: A Surveycited by 254
  12. 2019Task2Vec: Task Embedding for Meta-Learningcited by 197
29 more
  1. 2018Bayesian Model-Agnostic Meta-Learningcited by 194
  2. 2019Meta-Learning Representations for Continual Learningcited by 189
  3. 2018Recasting Gradient-Based Meta-Learning as Hierarchical Bayescited by 180
  4. 2019On the Convergence Theory of Gradient-Based Model-Agnostic Meta-Learning Algorithmscited by 132
  5. 2018Meta-Learning Probabilistic Inference For Predictioncited by 108
  6. 2021Provably Faster Algorithms for Bilevel Optimizationcited by 106
  7. 2019Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulationcited by 103
  8. 2019Meta-Learning without Memorizationcited by 92
  9. 2018ProMP: Proximal Meta-Policy Searchcited by 91
  10. 2020PACOH: Bayes-Optimal Meta-Learning with PAC-Guaranteescited by 85
  11. 2019Personalizing Dialogue Agents via Meta-Learningcited by 83
  12. 2018Unsupervised Learning via Meta-Learningcited by 81
  13. 2017Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theorycited by 76
  14. 2019Provable Guarantees for Gradient-Based Meta-Learningcited by 67
  15. 2021Towards Gradient-based Bilevel Optimization with Non-convex Followers and Beyondcited by 62
  16. 2020Solving Stochastic Compositional Optimization is Nearly as Easy as Solving Stochastic Optimizationcited by 51
  17. 2019Continuous Meta-Learning without Taskscited by 50
  18. 2021Generalization Bounds For Meta-Learning: An Information-Theoretic Analysiscited by 45
  19. 2021Generalization of Model-Agnostic Meta-Learning Algorithms: Recurring and Unseen Taskscited by 36
  20. 2020Information-Theoretic Generalization Bounds for Meta-Learning and Applicationscited by 32
  21. 2020How Important is the Train-Validation Split in Meta-Learning?cited by 30
  22. 2019Learning to Propagate for Graph Meta-Learningcited by 29
  23. 2021Generalization Bounds for Meta-Learning via PAC-Bayes and Uniform Stabilitycited by 29
  24. 2019Uncertainty in Model-Agnostic Meta-Learning using Variational Inferencecited by 23
  25. 2020On the Global Optimality of Model-Agnostic Meta-Learningcited by 21
  26. 2021Bridging the Gap Between Practice and PAC-Bayes Theory in Few-Shot Meta-Learningcited by 21
  27. 2020Transfer Meta-Learning: Information-Theoretic Bounds and Information Meta-Risk Minimizationcited by 14
  28. 2020Modeling and Optimization Trade-off in Meta-learningcited by 13
  29. 2020How Does the Task Landscape Affect MAML Performance?cited by 4

Built on by

Cited by year

Citing papers per year: 2022: 4; 2023: 5; 2024: 3; 2025: 1.

13 papers.

Sort:

  1. 2022Sharp-MAML: Sharpness-Aware Model-Agnostic Meta Learningcited by 57
  2. 2024Predicting Configuration Performance in Multiple Environments with Sequential Meta-learningcited by 17
  3. 2022Understanding Benign Overfitting in Gradient-Based Meta Learningcited by 11
  4. 2022Provable Generalization of Overparameterized Meta-learning Trained with SGDcited by 9
  5. 2023Bayes meets Bernstein at the Meta Level: an Analysis of Fast Rates in Meta-Learning with PAC-Bayescited by 8
  6. 2024Meta-Learning with Generalized Ridge Regression: High-dimensional Asymptotics, Optimality and Hyper-covariance Estimationcited by 5
  7. 2023Scalable Bayesian Meta-Learning through Generalized Implicit Gradientscited by 4
  8. 2023Theoretical Characterization of the Generalization Performance of Overfitted Meta-Learningcited by 4
  9. 2023Personalized Federated Learning via Amortized Bayesian Meta-Learningcited by 1
  10. 2022Hypernetwork approach to Bayesian MAMLcited by 0
  11. 2023A Hierarchical Bayesian Model for Deep Few-Shot Meta Learningcited by 0
  12. 2024Reducing Variance in Meta-Learning via Laplace Approximation for Regression Taskscited by 0
1 more
  1. 2025Domain-Generalization to Improve Learning in Meta-Learning Algorithmscited by 0