The DOI matches its DataCite record; 53 of the 55 references it cites were found in a registry or an index.
Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learning
Open accessAbstract
Although reinforcement learning methods can achieve impressive results in simulation, the real world presents two major challenges: generating samples is exceedingly expensive, and unexpected perturbations or unseen situations cause proficient but specialized policies to fail at test time. Given that it is impractical to train separate policies to accommodate all situations the agent may see in the real world, this work proposes to learn how to quickly and effectively adapt online to new tasks. To enable sample-efficient learning, we consider learning online adaptation in the context of model-based reinforcement learning. Our approach uses meta-learning to train a dynamics model prior such that, when combined with recent data, this prior can be rapidly adapted to the local context. Our experiments demonstrate online adaptation for continuous control tasks on both simulated and real-world agents. We first show simulated agents adapting their behavior online to novel terrains, crippled body parts, and highly-dynamic environments. We also illustrate the importance of incorporating online adaptation into autonomous agents that operate in the real world by applying our method to a real dynamic legged millirobot. We demonstrate the agent's learned ability to quickly adapt online to a missing leg, adjust to novel terrains and slopes, account for miscalibration or errors in pose estimation, and compensate for pulling payloads.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, the problem settings in these papers were offline as they had two separate phases, one for learning and the other for evaluation.”
From Loss of plasticity in deep continual learning (2024) · cites this paper as [122]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357
Continuous control with deep reinforcement learning
Lillicrap, T. P., Hunt, J. J., Pritzel, A., Heess, N., Erez, T., Tassa, Y., Silver, D., & Wierstra, D. (2015) · arXiv · doi:10.48550/arxiv.1509.02971 · Cited by 10,782
Mastering the game of Go without human knowledge
Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., Hubert, T., Baker, L., Lai, M., Bolton, A., et al. (2017) · Nature · doi:10.1038/nature24270 · Cited by 9,084
Åström, K. J., & Wittenmark, B. (1989) · Cited by 5,009
MuJoCo: A physics engine for model-based control
Todorov, E., Erez, T., & Tassa, Y. (2012) · 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2012.6386109 · Cited by 4,767
Trust region policy optimization
Schulman, J., Levine, S., Moritz, P., Jordan, M. I., & Abbeel, P. (2015) · arXiv · doi:10.48550/arxiv.1502.05477 · Cited by 4,578
Self-citationHuman-level concept learning through probabilistic program induction
Lake, B. M., Salakhutdinov, R., & Tenenbaum, J. B. (2015) · Science · doi:10.1126/science.aab3050 · Cited by 2,817
Optimization as a Model for Few-Shot Learning
Ravi, S., & Larochelle, H. (2017) · International Conference on Learning Representations · Cited by 2,766
End-to-End Training of Deep Visuomotor PoliciesPreprint
Levine, S., Finn, C., Darrell, T., & Abbeel, P. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1504.00702 · Cited by 2,641
Self-citationLearning to learn by gradient descent by gradient descent
Andrychowicz, M., Denil, M., Gomez, S., Hoffman, M. W., Pfau, D., Schaul, T., Shillingford, B., & de Freitas, N. (2016) · arXiv · doi:10.48550/arxiv.1606.04474 · Cited by 1,486
Research works printed but not linked: 6
- B. Baker, O. Gupta, N. Naik, and R. Raskar. Designing neural network architectures using reinforcement learning. arXiv preprint arXiv:1611.02167 , 2016. · arXiv
- K. Chua, R. Calandra, R. McAllister, and S. Levine. Deep reinforcement learning in a handful of trials using probabilistic dynamics models. arXiv preprint arXiv:1805.12114 , 2018. · arXiv
- M. P. Deisenroth, G. Neumann, J. Peters, et al. A survey on policy search for robotics. Foundations and Trends R © in Robotics , 2(1-2):1-142, 2013.
- Y. Duan, J. Schulman, X. Chen, P. L. Bartlett, I. Sutskever, and P. Abbeel. Rl$ˆ2$: Fast reinforcement learning via slow reinforcement learning. CoRR , abs/1611.02779, 2016. · arXiv
- C. Finn, P. Abbeel, and S. Levine. Model-agnostic meta-learning for fast adaptation of deep networks. CoRR , abs/1703.03400, 2017. · arXiv
- T. Munkhdalai and H. Yu. Meta networks. arXiv preprint arXiv:1703.00837 , 2017. · arXiv
Cited by
Sorted by citations
Deep Reinforcement Learning: A Survey
Wang, X., Wang, S., Liang, X., Zhao, D., Huang, J., Xu, X., Dai, B., & Miao, Q. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3207346 · Cited by 972
Meta-Learning in Neural Networks: A Survey
Hospedales, T. M., Antoniou, A., Micaelli, P., & Storkey, A. J. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3079209 · Cited by 886
Learning-Based Model Predictive Control: Toward Safe Learning in Control
Hewing, L., Wabersich, K. P., Menner, M., & Zeilinger, M. N. (2019) · Annual Review of Control Robotics and Autonomous Systems · doi:10.1146/annurev-control-090419-075625 · Cited by 849
How to train your robot with deep reinforcement learning: Lessons we have learned
Ibarz, J., Tan, J., Finn, C., Kalakrishnan, M., Pastor, P., & Levine, S. (2021) · The International Journal of Robotics Research · doi:10.1177/0278364920987859 · Cited by 570
Self-citationModel-based Reinforcement Learning: A Survey
Moerland, T. M., Broekens, J., Plaat, A., & Jonker, C. M. (2023) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000086 · Cited by 871
Metadata
- Authors
- Nagabandi, A., Clavera, I., Liu, S., Fearing, R. S., Abbeel, P., Levine, S., & Finn, C.
- Affiliations
- University of California, Berkeley (US)
- Venue
- arXiv
- Publisher
- arXiv
- Published
- 2018
- ISSN
- 2331-8422 (ISSN-L)
- arXiv
- 1803.11347
- OpenAlex
- W2903630557
- Cited by
- 370
- References
- 55
- Citations by year
- 2026: 3 · 2025: 25 · 2024: 47 · 2023: 46 · 2022: 61 · 2021: 100 · 2020: 65 · 2019: 18 · 2018: 5
- Type
- posted content
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Engineering › Control and Systems Engineering › Robotic Locomotion and Control
- Keywords
- meta-reinforcement learning, online adaptation, model-based reinforcement learning, meta-learning, dynamics model, sample efficiency, continuous control, legged robots, millirobots, terrain adaptation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
