Knowledge Reutilization in Meta-Reinforcement Learning
On the map: Deep reinforcement learning methods → Deep RL algorithms and representations
Abstract (arXiv)
Meta-reinforcement learning enables fast adaptation by extracting shared structure from related tasks, but existing end-to-end methods often couple task inference with embodiment-specific control. This coupling can obscure non-parametric task semantics, reduce sample efficiency, and limit cross-agent reuse. We propose a meta-knowledge reutilization framework that learns task-level knowledge on a dynamics-simplified agent and transfers it to heterogeneous agents. The framework uses a Bayesian non-parametric prior to organize latent task modes and a high-level policy to generate task-level magnitude guidance. To bridge reusable task knowledge with different embodiments, we introduce a semantic-magnitude interface and a lightweight temporal adaptor, which convert frozen meta-knowledge into temporally aligned subgoals for embodiment-specific low-level controllers. Experiments on multiple locomotion agents show that our framework reduces final-step tracking error by 94.75% -- 99.79% compared with recent state-of-the-art baselines and achieves comparable deployment performance with about 23.8% of their interaction data.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 18 of its 34 references that are in the map. First posted on arXiv 16 Jun 2026.
Builds on
18 of 34 references.
References · 18
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
- 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
- 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
- 2018On First-Order Meta-Learning Algorithmscited by 1,018
- 2016RL2: Fast Reinforcement Learning via Slow Reinforcement Learningcited by 503
- 2019Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variablescited by 389
- 2016Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoderscited by 220
- 2016Variational Deep Embedding: An Unsupervised and Generative Approach to Clusteringcited by 220
- 2019VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learningcited by 217
- 2022MetaMorph: Learning Universal Controllers with Transformerscited by 89
- 2022DeepDPM: Deep Clustering With an Unknown Number of Clusterscited by 52
- 2016Stick-Breaking Variational Autoencoderscited by 44
6 more
- 2020Hierarchically Decoupled Imitation for Morphological Transfercited by 31
- 2017Nonparametric Variational Auto-encoders for Hierarchical Representation Learningcited by 22
- 2023Recurrent Hypernetworks are Surprisingly Strong in Meta-RLcited by 16
- 2021Robust MAML: Prioritization task buffer with adaptive learning process for model-agnostic meta-learningcited by 10
- 2023Meta-Reinforcement Learning Based on Self-Supervised Task Representation Learningcited by 10
- 2024A Survey on Deep Clustering: From the Prior Perspectivecited by 9
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Meta-Learning and Meta-Reinforcement Learning -- Tracing the Path towards DeepMind's Adaptive Agent
- 2024DynaMITE-RL: A Dynamic Model for Improved Temporal Meta-Reinforcement Learning
- 2024Memory Sequence Length of Data Sampling Impacts the Adaptation of Meta-Reinforcement Learning Agents
- 2026Principled Fast and Meta Knowledge Learners for Continual Reinforcement Learning
- 2023Dream to Adapt: Meta Reinforcement Learning by Latent Context Imagination and MDP Imagination
- 2024AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers
- 2024Hierarchical Meta-Reinforcement Learning via Automated Macro-Action Discovery
- 2025Meta-Policy Reflexion: Reusable Reflective Memory and Rule Admissibility for Resource-Efficient LLM Agent