mlsanityData to 7 Oct 2026

Meta-Learning Probabilistic Inference For Prediction

arXiv 1805.09921 ↗First posted 24 May 2018stat.ML cs.LGcited by 108

Not on the map. Why?

Abstract (arXiv)

This paper introduces a new framework for data efficient and versatile learning. Specifically: 1) We develop ML-PIP, a general framework for Meta-Learning approximate Probabilistic Inference for Prediction. ML-PIP extends existing probabilistic interpretations of meta-learning to cover a broad class of methods. 2) We introduce VERSA, an instance of the framework employing a flexible and versatile amortization network that takes few-shot learning datasets as inputs, with arbitrary numbers of shots, and outputs a distribution over task-specific parameters in a single forward pass. VERSA substitutes optimization at test time with forward passes through inference networks, amortizing the cost of inference and relieving the need for second derivatives during training. 3) We evaluate VERSA on benchmark datasets where the method sets new state-of-the-art results, handles arbitrary numbers of shots, and for classification, arbitrary numbers of classes at train and test time. The power of the approach is then demonstrated through a challenging few-shot ShapeNet view reconstruction task.

Key facts. Cited by 108 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 May 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 28; 2022: 26; 2023: 19; 2024: 14; 2025: 15; 2026: 6.

108 papers.

Sort:

By area:

  1. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,337
  2. 2021Domain Generalization: A Surveycited by 760
  3. 2022Transformers learn in-context by gradient descentcited by 714
  4. 2021Transformers Can Do Bayesian Inferencecited by 338
  5. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Modelcited by 285
  6. 2021SMIL: Multimodal Learning with Severely Missing Modalitycited by 243
  7. 2022Self-Support Few-Shot Semantic Segmentationcited by 96
  8. 2022Tutorial on amortized optimizationcited by 83
  9. 2025Universal Actions for Enhanced Embodied Foundation Modelscited by 80
  10. 2021MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Imagescited by 78
  11. 2021Bayesian Learning via Neural Schr"odinger-F"ollmer Flowscited by 56
  12. 2022Bayesian Prompt Learning for Image-Language Model Generalizationcited by 51
38 more
  1. 2021Shallow Bayesian Meta Learning for Real-World Few-Shot Recognitioncited by 48
  2. 2023In-Context Learning Learns Label Relationships but Is Not Conventional Learningcited by 48
  3. 2021Property-Aware Relation Networks for Few-Shot Molecular Property Predictioncited by 41
  4. 2024Mixture of In-Context Prompters for Tabular PFNscited by 32
  5. 2021ORBIT: A Real-World Few-Shot Dataset for Teachable Object Recognitioncited by 29
  6. 2021Kernel Continual Learningcited by 27
  7. 2022Realistic Evaluation of Transductive Few-Shot Learningcited by 26
  8. 2022Learning to Generalize across Domains on Single Test Samplescited by 24
  9. 2022HyperShot: Few-Shot Learning by Kernel HyperNetworkscited by 24
  10. 2021The Internet of Federated Things (IoFT): A Vision for the Future and In-depth Survey of Data-driven Approaches for Federated Learningcited by 23
  11. 2021Bayesian Meta-Learning for Few-Shot Policy Adaptation Across Robotic Platformscited by 22
  12. 2021Meta-Learning Adversarial Domain Adaptation Network for Few-Shot Text Classificationcited by 18
  13. 2022What Matters For Meta-Learning Vision Regression Tasks?cited by 18
  14. 2022HyperMAML: Few-Shot Adaptation of Deep Models with Hypernetworkscited by 16
  15. 2021Variational Multi-Task Learning with Gumbel-Softmax Priorscited by 15
  16. 2021A contrastive rule for meta-learningcited by 14
  17. 2021Memory Efficient Meta-Learning with Large Imagescited by 14
  18. 2022Is Bayesian Model-Agnostic Meta Learning Better than Model-Agnostic Meta Learning, Provably?cited by 13
  19. 2023From system models to class models: An in-context learning paradigmcited by 13
  20. 2025Skill Expansion and Composition in Parameter Spacecited by 13
  21. 2021Non-Gaussian Gaussian Processes for Few-Shot Regressioncited by 12
  22. 2021Learning to Rectify for Robust Learning with Noisy Labelscited by 12
  23. 2023Patch-Prompt Aligned Bayesian Prompt Tuning for Vision-Language Modelscited by 12
  24. 2024Query-guided Prototype Evolution Network for Few-Shot Segmentationcited by 12
  25. 2021Adaptive Convolutions with Per-pixel Dynamic Filter Atomcited by 11
  26. 2023Model-Agnostic Meta-Learning for Multilingual Hate Speech Detectioncited by 11
  27. 2025Estimating Interventional Distributions with Uncertain Causal Graphs through Meta-Learningcited by 11
  28. 2021Information-Theoretic Analysis of Epistemic Uncertainty in Bayesian Meta-learningcited by 10
  29. 2022Adaptive Dimension Reduction and Variational Inference for Transductive Few-Shot Classificationcited by 10
  30. 2022Evidential Conditional Neural Processescited by 10
  31. 2022Multidimensional Belief Quantification for Label-Efficient Meta-Learningcited by 9
  32. 2023A Simple Yet Effective Strategy to Robustify the Meta Learning Paradigmcited by 9
  33. 2021Bayesian Embeddings for Few-Shot Open World Recognitioncited by 8
  34. 2022Hidden Parameter Recurrent State Space Models For Changing Dynamics Scenarioscited by 8
  35. 2023Episodic Multi-Task Learning with Heterogeneous Neural Processescited by 8
  36. 2023Hacking Task Confounder in Meta-Learningcited by 8
  37. 2024Learning to Continually Learn with the Bayesian Principlecited by 8
  38. 2021Few-Shot Video Object Detectioncited by 7

50 most cited of 108 All 108 (JSON)