Abstract
Offline Meta Reinforcement Learning (OMRL) aims to learn transferable knowledge from offline datasets to enhance the learning process for new target tasks. Context-based Reinforcement Learning (RL) adopts a context encoder to expediently adapt the agent to new tasks by inferring the task representation, and then adjusting the policy based on this inferred representation. In this work, we focus on context-based OMRL, specifically on the challenge of learning task representation for OMRL. We conduct experiments that demonstrate that the context encoder trained on offline datasets might encounter distribution shift between the contexts used for training and testing. To overcome this problem, we present a hard-sampling-based strategy to train a robust task context encoder. Our experimental findings on diverse continuous control tasks reveal that utilizing our approach yields more robust task representations and better testing performance in terms of accumulated returns compared to baseline methods. Our code is available at https://github.com/ZJLAB-AMMI/HS-OMRL.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bengio, Y., Bengio, S., Cloutier, J.: Learning a synaptic learning rule. Citeseer (1990)
Berner, C., et al.: Dota 2 with large scale deep reinforcement learning. arXiv preprint arXiv:1912.06680 (2019)
Chen, T., Kornblith, S., Norouzi, M., Hinton, G.: A simple framework for contrastive learning of visual representations. In: International Conference on Machine Learning, pp. 1597–1607. PMLR (2020)
Dorfman, R., Shenfeld, I., Tamar, A.: Offline meta reinforcement learning–identifiability challenges and effective data collection strategies. Adv. Neural Inf. Process. Syst. 34, 4607–4618 (2021)
Duan, Y., Schulman, J., Chen, X., Bartlett, P.L., Sutskever, I., Abbeel, P.: Rl2: fast reinforcement learning via slow reinforcement learning. arXiv preprint arXiv:1611.02779 (2016)
Finn, C., Abbeel, P., Levine, S.: Model-agnostic meta-learning for fast adaptation of deep networks. In: International Conference on Machine Learning, pp. 1126–1135. PMLR (2017)
Gottesman, O., et al.: Guidelines for reinforcement learning in healthcare. Nat. Med. 25(1), 16–18 (2019)
Haarnoja, T., Zhou, A., Abbeel, P., Levine, S.: Soft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor. In: International Conference on Machine Learning, pp. 1861–1870. PMLR (2018)
He, K., Fan, H., Wu, Y., Xie, S., Girshick, R.: Momentum contrast for unsupervised visual representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9729–9738 (2020)
Hospedales, T., Antoniou, A., Micaelli, P., Storkey, A.: Meta-learning in neural networks: a survey. IEEE Trans. Pattern Anal. Mach. Intell. 44(9), 5149–5169 (2021)
Huang, W., Yi, M., Zhao, X.: Towards the generalization of contrastive self-supervised learning. arXiv preprint arXiv:2111.00743 (2021)
Humplik, J., Galashov, A., Hasenclever, L., Ortega, P.A., Teh, Y.W., Heess, N.: Meta reinforcement learning as task inference. arXiv preprint arXiv:1905.06424 (2019)
Khosla, P., et al.: Supervised contrastive learning. In: Advances in Neural Information Processing Systems 33, pp. 18661–18673 (2020)
Kostrikov, I., Nair, A., Levine, S.: Offline reinforcement learning with implicit Q-learning. arXiv preprint arXiv:2110.06169 (2021)
Levine, S., Kumar, A., Tucker, G., Fu, J.: Offline reinforcement learning: tutorial, review, and perspectives on open problems. arXiv preprint arXiv:2005.01643 (2020)
Li, L., Huang, Y., Chen, M., Luo, S., Luo, D., Huang, J.: Provably improved context-based offline meta-RL with attention and contrastive learning. arXiv preprint arXiv:2102.10774 (2021)
Li, L., Yang, R., Luo, D.: Focal: efficient fully-offline meta-reinforcement learning via distance metric learning and behavior regularization. arXiv preprint arXiv:2010.01112 (2020)
Mitchell, E., Rafailov, R., Peng, X.B., Levine, S., Finn, C.: Offline meta-reinforcement learning with advantage weighting. In: International Conference on Machine Learning, pp. 7780–7791. PMLR (2021)
Mnih, V., et al.: Human-level control through deep reinforcement learning. Nature 518(7540), 529–533 (2015)
Oord, A.v.d., Li, Y., Vinyals, O.: Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748 (2018)
Peng, X.B., Kumar, A., Zhang, G., Levine, S.: Advantage-weighted regression: simple and scalable off-policy reinforcement learning. arXiv preprint arXiv:1910.00177 (2019)
Prudencio, R.F., Maximo, M.R., Colombini, E.L.: A survey on offline reinforcement learning: taxonomy, review, and open problems. arXiv preprint arXiv:2203.01387 (2022)
Rakelly, K., Zhou, A., Finn, C., Levine, S., Quillen, D.: Efficient off-policy meta-reinforcement learning via probabilistic context variables. In: International Conference on Machine Learning, pp. 5331–5340. PMLR (2019)
Robinson, J., Chuang, C.Y., Sra, S., Jegelka, S.: Contrastive learning with hard negative samples. arXiv preprint arXiv:2010.04592 (2020)
Schmidhuber, J.: Evolutionary principles in self-referential learning, or on learning how to learn: the meta-meta-... hook. Ph.D. thesis, Technische Universität München (1987)
Silver, D., et al.: Mastering chess and shogi by self-play with a general reinforcement learning algorithm. arXiv preprint arXiv:1712.01815 (2017)
Wang, F., Liu, H.: Understanding the behaviour of contrastive loss. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2495–2504 (2021)
Wang, J.X., et al.: Learning to reinforcement learn. arXiv preprint arXiv:1611.05763 (2016)
Wang, T., Isola, P.: Understanding contrastive representation learning through alignment and uniformity on the hypersphere. In: International Conference on Machine Learning, pp. 9929–9939. PMLR (2020)
Wu, Y., Tucker, G., Nachum, O.: Behavior regularized offline reinforcement learning. arXiv preprint arXiv:1911.11361 (2019)
Yuan, H., Lu, Z.: Robust task representations for offline meta-reinforcement learning via contrastive learning. In: International Conference on Machine Learning, pp. 25747–25759. PMLR (2022)
Zhou, M., et al.: Smarts: scalable multi-agent reinforcement learning training school for autonomous driving. arXiv preprint arXiv:2010.09776 (2020)
Zintgraf, L., et al.: VariBAD: a very good method for Bayes-adaptive deep RL via meta-learning. arXiv preprint arXiv:1910.08348 (2019)
Zhao, C., Zhou, Z., Liu, B.: On context distribution shift in task representation learning for offline meta RL. arXiv preprint arXiv: 2304.00354 (2023)
Acknowledgement
This work was supported by Exploratory Research Project (No. 2022RC0AN02) of Zhejiang Lab.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhao, C., Zhou, Z., Liu, B. (2023). On Context Distribution Shift in Task Representation Learning for Online Meta RL. In: Huang, DS., Premaratne, P., Jin, B., Qu, B., Jo, KH., Hussain, A. (eds) Advanced Intelligent Computing Technology and Applications. ICIC 2023. Lecture Notes in Computer Science(), vol 14090. Springer, Singapore. https://doi.org/10.1007/978-981-99-4761-4_52
Download citation
DOI: https://doi.org/10.1007/978-981-99-4761-4_52
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4760-7
Online ISBN: 978-981-99-4761-4
eBook Packages: Computer ScienceComputer Science (R0)