File Download

There are no files associated with this item.

Supplementary

Conference Paper: Model-based reinforcement learning via imagination with derived memory

TitleModel-based reinforcement learning via imagination with derived memory
Authors
KeywordsModel-based Reinforcement Learning
Policy Robustness
Visual Control Task
Derived Memory
World Models
Issue Date2021
PublisherNeural Information Processing Systems Foundation.
Citation
35th Conference on Neural Information Processing Systems (NeurIPS 2021) (Virtual), December 6-14, 2021. In Advances In Neural Information Processing Systems: 35th conference on neural information processing systems (NeurIPS 2021), p. 9493-9505 How to Cite?
AbstractModel-based reinforcement learning aims to improve the sample efficiency of policy learning by modeling the dynamics of the environment. Recently, the latent dynamics model is further developed to enable fast planning in a compact space. It summarizes the high-dimensional experiences of an agent, which mimics the memory function of humans. Learning policies via imagination with the latent model shows great potential for solving complex tasks. However, only considering memories from the true experiences in the process of imagination could limit its advantages. Inspired by the memory prosthesis proposed by neuroscientists, we present a novel model-based reinforcement learning framework called Imagining with Derived Memory (IDM). It enables the agent to learn policy from enriched diverse imagination with prediction-reliability weight, thus improving sample efficiency and policy robustness. Experiments on various high-dimensional visual control tasks in the DMControl benchmark demonstrate that IDM outperforms previous state-of-the-art methods in terms of policy robustness and further improves the sample efficiency of the model-based method.
Persistent Identifierhttp://hdl.handle.net/10722/315861

 

DC FieldValueLanguage
dc.contributor.authorMu, Y-
dc.contributor.authorZhuang, Y-
dc.contributor.authorWang, B-
dc.contributor.authorZhu, G-
dc.contributor.authorLiu, W-
dc.contributor.authorChen, J-
dc.contributor.authorLuo, P-
dc.contributor.authorLi, SE-
dc.contributor.authorZhang, C-
dc.contributor.authorHao, J-
dc.date.accessioned2022-08-19T09:05:47Z-
dc.date.available2022-08-19T09:05:47Z-
dc.date.issued2021-
dc.identifier.citation35th Conference on Neural Information Processing Systems (NeurIPS 2021) (Virtual), December 6-14, 2021. In Advances In Neural Information Processing Systems: 35th conference on neural information processing systems (NeurIPS 2021), p. 9493-9505-
dc.identifier.urihttp://hdl.handle.net/10722/315861-
dc.description.abstractModel-based reinforcement learning aims to improve the sample efficiency of policy learning by modeling the dynamics of the environment. Recently, the latent dynamics model is further developed to enable fast planning in a compact space. It summarizes the high-dimensional experiences of an agent, which mimics the memory function of humans. Learning policies via imagination with the latent model shows great potential for solving complex tasks. However, only considering memories from the true experiences in the process of imagination could limit its advantages. Inspired by the memory prosthesis proposed by neuroscientists, we present a novel model-based reinforcement learning framework called Imagining with Derived Memory (IDM). It enables the agent to learn policy from enriched diverse imagination with prediction-reliability weight, thus improving sample efficiency and policy robustness. Experiments on various high-dimensional visual control tasks in the DMControl benchmark demonstrate that IDM outperforms previous state-of-the-art methods in terms of policy robustness and further improves the sample efficiency of the model-based method.-
dc.languageeng-
dc.publisherNeural Information Processing Systems Foundation.-
dc.relation.ispartofAdvances In Neural Information Processing Systems: 35th conference on neural information processing systems (NeurIPS 2021)-
dc.subjectModel-based Reinforcement Learning-
dc.subjectPolicy Robustness-
dc.subjectVisual Control Task-
dc.subjectDerived Memory-
dc.subjectWorld Models-
dc.titleModel-based reinforcement learning via imagination with derived memory-
dc.typeConference_Paper-
dc.identifier.emailLuo, P: pluo@hku.hk-
dc.identifier.authorityLuo, P=rp02575-
dc.identifier.hkuros335597-
dc.identifier.spage9493-
dc.identifier.epage9505-
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats