mlsanityData to 14 Oct 2026

Dream to Adapt: Meta Reinforcement Learning by Latent Context Imagination and MDP Imagination

arXiv 2311.06673 ↗First posted 11 Nov 2023cs.LG cs.AI cs.ROcited by 6

Not on the map. Why?

Abstract (arXiv)

Meta reinforcement learning (Meta RL) has been amply explored to quickly learn an unseen task by transferring previously learned knowledge from similar tasks. However, most state-of-the-art algorithms require the meta-training tasks to have a dense coverage on the task distribution and a great amount of data for each of them. In this paper, we propose MetaDreamer, a context-based Meta RL algorithm that requires less real training tasks and data by doing meta-imagination and MDP-imagination. We perform meta-imagination by interpolating on the learned latent context space with disentangled properties, as well as MDP-imagination through the generative world model where physical knowledge is added to plain VAE networks. Our experiments with various benchmarks show that MetaDreamer outperforms existing approaches in data efficiency and interpolated generalization.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 35 of its 60 references that are in the map. First posted on arXiv 11 Nov 2023.

Builds on

35 of 60 references.

References · 35

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  2. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  3. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  4. 2018On First-Order Meta-Learning Algorithmscited by 1,018
  5. 2020Mastering Atari with Discrete World Modelscited by 1,007
  6. 2018Self-Attention Generative Adversarial Networkscited by 809
  7. 2016RL2: Fast Reinforcement Learning via Slow Reinforcement Learningcited by 503
  8. 2019Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variablescited by 389
  9. 2016Learning to reinforcement learncited by 377
  10. 2016Benchmarking Deep Reinforcement Learning for Continuous Controlcited by 371
  11. 2017A Semantic Loss Function for Deep Learning with Symbolic Knowledgecited by 270
  12. 2016Data Programming: Creating Large Training Sets, Quicklycited by 246
23 more
  1. 2017Physics-guided Neural Networks (PGNN): An Application in Lake Temperature Modelingcited by 131
  2. 2018Meta-Reinforcement Learning of Structured Exploration Strategiescited by 130
  3. 2019Transfer learning enhanced physics informed neural network for phase-field modeling of fracturecited by 116
  4. 2016Label-Free Supervision of Neural Networks with Physics and Domain Knowledgecited by 74
  5. 2016Understanding data augmentation for classification: when to warp?cited by 74
  6. 2019Improving Generalization in Meta Reinforcement Learning using Learned Objectivescited by 68
  7. 2021Physics-Integrated Variational Autoencoders for Robust and Interpretable Generative Modelingcited by 59
  8. 2020Improving Generalization in Meta-learning via Task Augmentationcited by 55
  9. 2021One-Shot Transfer Learning of Physics-Informed Neural Networkscited by 49
  10. 2020Data Augmentation for Meta-Learningcited by 38
  11. 2020Model-based Adversarial Meta-Reinforcement Learningcited by 33
  12. 2022DreamingV2: Reinforcement Learning with Discrete World Models without Reconstructioncited by 29
  13. 2019Guided Meta-Policy Searchcited by 26
  14. 2020Meta-Reinforcement Learning Robust to Distributional Shift via Model Identification and Experience Relabelingcited by 26
  15. 2020Unsupervised Meta-Learning through Latent-Space Interpolation in Generative Modelscited by 26
  16. 2017Beating Atari with Natural Language Guided Reinforcement Learningcited by 24
  17. 2020A theory of independent mechanisms for extrapolation in generative modelscited by 17
  18. 2018Physics-informed deep generative modelscited by 16
  19. 2021Improving Generalization in Meta-RL with Imaginary Tasks from Latent Dynamics Mixturecited by 12
  20. 2021Disentanglement Analysis with Partial Information Decompositioncited by 9
  21. 2018Adversarial Constraint Learning for Structured Predictioncited by 3
  22. 2021Quick Learner Automated Vehicle Adapting its Roadmanship to Varying Traffic Cultures with Meta Reinforcement Learningcited by 3
  23. 2020Local Nonparametric Meta-Learningcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 2; 2026: 2.

6 papers.

  1. 2024Dreaming of Many Worlds: Learning Contextual World Models Aids Zero-Shot Generalizationcited by 16
  2. 2024Inferring Behavior-Specific Context Improves Zero-Shot Generalization in Reinforcement Learningcited by 5
  3. 2025Task-Aware Virtual Training: Enhancing Generalization in Meta-Reinforcement Learning for Out-of-Distribution Taskscited by 3
  4. 2026Disentangling Dynamical Systems: Causal Representation Learning Meets Local Sparse Attentioncited by 3
  5. 2025cMALC-D: Contextual Multi-Agent LLM-Guided Curriculum Learning with Diversity-Based Context Blendingcited by 2
  6. 2026Improving Generalization and Robustness in Offline Reinforcement Learning via Boundary-Aware Data Augmentationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Meta-Learning and Meta-Reinforcement Learning -- Tracing the Path towards DeepMind's Adaptive Agent
  2. 2024AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers
  3. 2026Behavior-Invariant Task Representation Learning with Transformer-based World Models for Offline Meta-Reinforcement Learning
  4. 2026Knowledge Reutilization in Meta-Reinforcement Learning
  5. 2021Offline Meta-Reinforcement Learning with Online Self-Supervision
  6. 2024Meta-DT: Offline Meta-RL as Conditional Sequence Modeling with World Model Disentanglement
  7. 2024DynaMITE-RL: A Dynamic Model for Improved Temporal Meta-Reinforcement Learning
  8. 2024Hierarchical Transformers are Efficient Meta-Reinforcement Learners