mlsanityData to 7 Oct 2026

Meta-learning of Sequential Strategies

arXiv 1905.03030 ↗First posted 8 May 2019cs.LG cs.AI stat.MLcited by 68

Not on the map. Why?

Abstract (arXiv)

In this report we review memory-based meta-learning as a tool for building sample-efficient strategies that learn from past experience to adapt to any task within a target class. Our goal is to equip the reader with the conceptual foundations of this tool for building new, scalable agents that operate on broad domains. To do so, we present basic algorithmic templates for building near-optimal predictors and reinforcement learners which behave as if they had a probabilistic model that allowed them to efficiently exploit task structure. Furthermore, we recast memory-based meta-learning within a Bayesian framework, showing that the meta-learned strategies are near-optimal because they amortize Bayes-filtered data, where the adaptation is implemented in the memory dynamics as a state-machine of sufficient statistics. Essentially, memory-based meta-learning translates the hard problem of probabilistic sequential inference into a regression problem.

Key facts. Cited by 68 papers in the map (data through 7 Oct 2026). First posted on arXiv 8 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 13; 2022: 13; 2023: 10; 2024: 17; 2025: 8; 2026: 7.

68 papers.

Sort:

By area:

  1. 2021Transformers Can Do Bayesian Inferencecited by 338
  2. 2021MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scalecited by 226
  3. 2023Human-Timescale Adaptation in an Open-Ended Task Spacecited by 136
  4. 2022General-Purpose In-Context Learning by Meta-Learning Transformerscited by 102
  5. 2024Transformers represent belief state geometry in their residual streamcited by 73
  6. 2021Meta-Thompson Samplingcited by 62
  7. 2023Meta-in-context learning in large language modelscited by 59
  8. 2022Transformers are Meta-Reinforcement Learnerscited by 39
  9. 2023AMAGO: Scalable In-Context Reinforcement Learning for Adaptive Agentscited by 36
  10. 2021Alchemy: A benchmark and analysis toolkit for meta-reinforcement learning agentscited by 35
  11. 2021Hierarchical Bayesian Banditscited by 33
  12. 2024The broader spectrum of in-context learningcited by 33
38 more
  1. 2024LMAct: A Benchmark for In-Context Imitation Learning with Long Multimodal Demonstrationscited by 32
  2. 2021Deep Interactive Bayesian Reinforcement Learning via Meta-Learningcited by 29
  3. 2024Learning Universal Predictorscited by 29
  4. 2022Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machinescited by 20
  5. 2024Lossless data compression by large modelscited by 20
  6. 2024Retrieval-Augmented Decision Transformer: External Memory for In-context RLcited by 20
  7. 2021Metalearning Linear Bandits by Prior Updatecited by 15
  8. 2022On the link between conscious function and general intelligence in humans and machinescited by 15
  9. 2023Memory-Based Meta-Learning on Non-Stationary Distributionscited by 14
  10. 2022Modeling Human Exploration Through Resource-Rational Reinforcement Learningcited by 13
  11. 2023In-Context Learning Dynamics with Random Binary Sequencescited by 13
  12. 2024In-context learning agents are asymmetric belief updaterscited by 13
  13. 2024Language Models are Symbolic Learners in Arithmeticcited by 11
  14. 2022Learning Robust Real-Time Cultural Transmission without Human Datacited by 9
  15. 2023ContraBAR: Contrastive Bayes-Adaptive Deep RLcited by 9
  16. 2023Social Contract AI: Aligning AI Assistants with Implicit Group Normscited by 9
  17. 2022Simplex Neural Population Learning: Any-Mixture Bayes-Optimality in Symmetric Zero-sum Gamescited by 8
  18. 2021Model-Free Risk-Sensitive Reinforcement Learningcited by 7
  19. 2024In-context learning and Occam's razorcited by 7
  20. 2022Meta Reinforcement Learning with Finite Training Tasks -- a Density Estimation Approachcited by 6
  21. 2025Understanding Prompt Tuning and In-Context Learning via Meta-Learningcited by 6
  22. 2021Unifying Gradient Estimators for Meta-Reinforcement Learning via Off-Policy Evaluationcited by 5
  23. 2021Biased Gradient Estimate with Drastic Variance Reduction for Meta Reinforcement Learningcited by 5
  24. 2023MONGOOSE: Path-wise Smooth Bayesian Optimisation via Meta-learningcited by 5
  25. 2024Transformers for Supervised Online Continual Learningcited by 5
  26. 2024Towards Understanding the Relationship between In-context Learning and Compositional Generalizationcited by 5
  27. 2025Next-token pretraining implies in-context learningcited by 5
  28. 2022Meta-Reinforcement Learning with Self-Modifying Networkscited by 4
  29. 2022Planning to the Information Horizon of BAMDPs via Epistemic State Abstractioncited by 4
  30. 2025Why is prompting hard? Understanding prompts on binary sequence predictorscited by 4
  31. 2021What is Going on Inside Recurrent Meta Reinforcement Learning Agents?cited by 3
  32. 2024Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networkscited by 3
  33. 2025A "good regulator theorem" for embodied agentscited by 3
  34. 2026Distinct mechanisms underlying in-context learning in transformerscited by 3
  35. 2026Position: agentic AI orchestration should be Bayes-consistentcited by 3
  36. 2021Embedding Adaptation is Still Needed for Few-Shot Learningcited by 2
  37. 2022Online Meta-Learning in Adversarial Multi-Armed Banditscited by 2
  38. 2023Learning Functional Transductioncited by 2

50 most cited of 68 All 68 (JSON)