mlsanityData to 14 Oct 2026

Meta-SGD: Learning to Learn Quickly for Few-Shot Learning

arXiv 1707.09835 ↗First posted 31 Jul 2017cs.LGcited by 379

Not on the map. Why?

Abstract (arXiv)

Few-shot learning is challenging for learning algorithms that learn each task in isolation and from scratch. In contrast, meta-learning learns from many related tasks a meta-learner that can learn a new task more accurately and faster with fewer examples, where the choice of meta-learners is crucial. In this paper, we develop Meta-SGD, an SGD-like, easily trainable meta-learner that can initialize and adapt any differentiable learner in just one step, on both supervised learning and reinforcement learning. Compared to the popular meta-learner LSTM, Meta-SGD is conceptually simpler, easier to implement, and can be learned more efficiently. Compared to the latest meta-learner MAML, Meta-SGD has a much higher capacity by learning to learn not just the learner initialization, but also the learner update direction and learning rate, all in a single meta-learning process. Meta-SGD shows highly competitive performance for few-shot learning on regression, classification, and reinforcement learning.

Key facts. Cited by 379 papers in the map (data through 14 Oct 2026). First posted on arXiv 31 Jul 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 106; 2022: 86; 2023: 73; 2024: 50; 2025: 37; 2026: 27.

379 papers.

Sort:

By area:

  1. 2022Transformers learn in-context by gradient descentcited by 732
  2. 2021Fast Model Editing at Scalecited by 498
  3. 2021Personalized Federated Learning using Hypernetworkscited by 212
  4. 2021Learning to Optimize: A Primer and A Benchmarkcited by 208
  5. 2022From data to functa: Your data point is a function and you can treat it like onecited by 187
  6. 2021Free Lunch for Few-shot Learning: Distribution Calibrationcited by 165
  7. 2022A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunitiescited by 163
  8. 2021Investigating Bi-Level Optimization for Learning and Vision from a Unified Perspective: A Survey and Beyondcited by 153
  9. 2021DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detectioncited by 106
  10. 2023Data Distillation: A Surveycited by 87
  11. 2023Operator Learning with Neural Fields: Tackling PDEs on General Geometriescited by 84
  12. 2021MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Imagescited by 78
38 more
  1. 2021Personalized Federated Learning with Gaussian Processescited by 76
  2. 2021Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learningcited by 72
  3. 2021Learning Normal Dynamics in Videos with Meta Prototype Networkcited by 71
  4. 2021MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaptioncited by 65
  5. 2023Advances and Challenges in Meta-Learning: A Technical Reviewcited by 65
  6. 2022Sharp-MAML: Sharpness-Aware Model-Agnostic Meta Learningcited by 58
  7. 2022Self-Destructing Models: Increasing the Costs of Harmful Dual Uses of Foundation Modelscited by 57
  8. 2021Learning Meta-class Memory for Few-Shot Semantic Segmentationcited by 56
  9. 2023Learning with Fantasy: Semantic-Aware Virtual Contrastive Constraint for Few-Shot Class-Incremental Learningcited by 55
  10. 2021Rectifying the Shortcut Learning of Background for Few-Shot Learningcited by 51
  11. 2021StyleMeUp: Towards Style-Agnostic Sketch-Based Image Retrievalcited by 50
  12. 2022Generalizing to New Physical Systems via Context-Informed Dynamics Modelcited by 50
  13. 2022Hybrid Relation Guided Set Matching for Few-shot Action Recognitioncited by 50
  14. 2021Meta-Learning with Task-Adaptive Loss Function for Few-Shot Learningcited by 49
  15. 2021Shallow Bayesian Meta Learning for Real-World Few-Shot Recognitioncited by 48
  16. 2023A Closer Look at Few-shot Classification Againcited by 46
  17. 2021Meta-Learning with Fewer Tasks through Task Interpolationcited by 45
  18. 2022Hypernetworks in Meta-Reinforcement Learningcited by 42
  19. 2021Learning Dynamic Alignment via Meta-filter for Few-shot Learningcited by 41
  20. 2021How Fine-Tuning Allows for Effective Meta-Learningcited by 41
  21. 2023CLIP-guided Prototype Modulating for Few-shot Action Recognitioncited by 41
  22. 2022Generating Representative Samples for Few-Shot Classificationcited by 39
  23. 2023MoLo: Motion-augmented Long-short Contrastive Learning for Few-shot Action Recognitioncited by 37
  24. 2021Generalization of Model-Agnostic Meta-Learning Algorithms: Recurring and Unseen Taskscited by 36
  25. 2021Controllable Semantic Parsing via Retrieval Augmentationcited by 35
  26. 2022Sketch3T: Test-Time Training for Zero-Shot SBIRcited by 34
  27. 2021Learning where to learn: Gradient sparsity in meta and continual learningcited by 33
  28. 2021Meta-Teacher For Face Anti-Spoofingcited by 33
  29. 2021A Metalearning Approach for Physics-Informed Neural Networks (PINNs): Application to Parameterized PDEscited by 32
  30. 2022Few-Shot Classification with Contrastive Learningcited by 31
  31. 2024AutoLoRA: Automatically Tuning Matrix Ranks in Low-Rank Adaptation Based on Meta Learningcited by 31
  32. 2021Dual-Awareness Attention for Few-Shot Object Detectioncited by 29
  33. 2021A General Descent Aggregation Framework for Gradient-based Bi-level Optimizationcited by 28
  34. 2021Deep Reinforcement Learning Versus Evolution Strategies: A Comparative Surveycited by 28
  35. 2021MetaHTR: Towards Writer-Adaptive Handwritten Text Recognitioncited by 27
  36. 2023Ensuring User-side Fairness in Dynamic Recommender Systemscited by 27
  37. 2021Binocular Mutual Learning for Improving Few-shot Classificationcited by 26
  38. 2022Gradient-based Bi-level Optimization for Deep Learning: A Surveycited by 26

50 most cited of 379 All 379 (JSON)