DC Field | Value | Language |
---|---|---|
dc.contributor.author | Hong, Sunghoon | ko |
dc.contributor.author | Yoon, Deunsol | ko |
dc.contributor.author | Kim, Kee-Eung | ko |
dc.date.accessioned | 2023-09-15T05:00:21Z | - |
dc.date.available | 2023-09-15T05:00:21Z | - |
dc.date.created | 2023-09-15 | - |
dc.date.issued | 2022-04 | - |
dc.identifier.citation | 10th International Conference on Learning Representations, ICLR 2022 | - |
dc.identifier.uri | http://hdl.handle.net/10203/312663 | - |
dc.description.abstract | Modular Reinforcement Learning, where the agent is assumed to be morphologically structured as a graph, for example composed of limbs and joints, aims to learn a policy that is transferable to a structurally similar but different agent. Compared to traditional Multi-Task Reinforcement Learning, this promising approach allows us to cope with inhomogeneous tasks where the state and action space dimensions differ across tasks. Graph Neural Networks are a natural model for representing the pertinent policies, but a recent work has shown that their multi-hop message passing mechanism is not ideal for conveying important information to other modules and thus a transformer model without morphological information was proposed. In this work, we argue that the morphological information is still very useful and propose a transformer policy model that effectively encodes such information. Specifically, we encode the morphological information in terms of the traversal-based positional embedding and the graph-based relational embedding. We empirically show that the morphological information is crucial for modular reinforcement learning, substantially outperforming prior state-of-the-art methods on multi-task learning as well as transfer learning settings with different state and action space dimensions. | - |
dc.language | English | - |
dc.publisher | International Conference on Learning Representations, ICLR | - |
dc.title | STRUCTURE-AWARE TRANSFORMER POLICY FOR INHOMOGENEOUS MULTI-TASK REINFORCEMENT LEARNING | - |
dc.type | Conference | - |
dc.identifier.scopusid | 2-s2.0-85143666294 | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | 10th International Conference on Learning Representations, ICLR 2022 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Virtual | - |
dc.contributor.localauthor | Kim, Kee-Eung | - |
dc.contributor.nonIdAuthor | Hong, Sunghoon | - |
dc.contributor.nonIdAuthor | Yoon, Deunsol | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.