mlsanityData to 7 Oct 2026

Binomial Gradient-Based Meta-Learning for Enhanced Meta-Gradient Estimation

arXiv 2604.13263 ↗First posted 14 Apr 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Meta-learning offers a principled framework leveraging task-invariant priors from related tasks, with which task-specific models can be fine-tuned on downstream tasks, even with limited data records. Gradient-based meta-learning (GBML) relies on gradient descent (GD) to adapt the prior to a new task. Albeit effective, these methods incur high computational overhead that scales linearly with the number of GD steps. To enhance efficiency and scalability, existing methods approximate the gradient of prior parameters (meta-gradient) via truncated backpropagation, yet suffer large approximation errors. Targeting accurate approximation, this work puts forth binomial GBML (BinomGBML), which relies on a truncated binomial expansion for meta-gradient estimation. This novel expansion endows more information in the meta-gradient estimation via efficient parallel computation. As a running paradigm applied to model-agnostic meta-learning (MAML), the resultant BinomMAML provably enjoys error bounds that not only improve upon existing approaches, but also decay super-exponentially under mild conditions. Numerical tests corroborate the theoretical analysis and showcase boosted performance with slightly increased computational overhead.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 42 references that are in the map. First posted on arXiv 14 Apr 2026.

Builds on

30 of 42 references.

References · 30

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  5. 2016Matching Networks for One Shot Learningcited by 2,246
  6. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  7. 2016Learning to learn by gradient descent by gradient descentcited by 616
  8. 2019Meta-Learning with Implicit Gradientscited by 523
  9. 2018Meta-Learning with Latent Embedding Optimizationcited by 476
  10. 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
  11. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  12. 2018Lipschitz regularity of deep neural networks: analysis and efficient estimationcited by 413
18 more
  1. 2018TADAM: Task dependent adaptive metric for improved few-shot learningcited by 408
  2. 2017Meta-SGD: Learning to Learn Quickly for Few-Shot Learningcited by 377
  3. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  4. 2018Approximation Methods for Bilevel Programmingcited by 231
  5. 2020Bilevel Optimization: Convergence Analysis and Enhanced Designcited by 219
  6. 2018Bayesian Model-Agnostic Meta-Learningcited by 194
  7. 2018Truncated Back-propagation for Bilevel Optimizationcited by 187
  8. 2019Adaptive Gradient-Based Meta-Learning Methodscited by 183
  9. 2018Recasting Gradient-Based Meta-Learning as Hierarchical Bayescited by 180
  10. 2021Meta-learning via Language Model In-context Tuningcited by 163
  11. 2019A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithmscited by 136
  12. 2017Non-convex Optimization for Machine Learningcited by 132
  13. 2019On the Convergence Theory of Gradient-Based Model-Agnostic Meta-Learning Algorithmscited by 132
  14. 2019Meta-Curvaturecited by 70
  15. 2020learn2learn: A Library for Meta-Learning Researchcited by 69
  16. 2021On the Convergence Theory for Hessian-Free Bilevel Algorithmscited by 33
  17. 2022Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learningcited by 22
  18. 2023Scalable Bayesian Meta-Learning through Generalized Implicit Gradientscited by 4

Built on by

Not cited yet.