Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385488822> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4385488822 abstract "Multi-task reinforcement learning (MTRL) is a technique to train multiple tasks simultaneously, where previous works usually train a single model to solve different tasks by sharing parameters across various tasks. However, these methods are faced with inter-task interference since what parameters should be shared across tasks is not addressed, dramatically reducing learning efficiency. To solve these problems, we propose a novel MTRL framework called Task-Specific feature Selector and Scheduler (T3S), which consists of two components: a feature selector and a task scheduler. Specifically, the feature selectors employ hypernetworks to construct task-specific soft masks, which can be applied by globally shared representation to construct task-specific features. The task scheduler selects tasks for learning through two metrics, where the selection probability is inversely proportional to task progress (e.g., success rate) and task learning speed. Experimental results show that T3S consistently outperforms the state-of-the-art MTRL algorithms on various robotics manipulation tasks." @default.
- W4385488822 created "2023-08-03" @default.
- W4385488822 creator A5013972839 @default.
- W4385488822 creator A5017719253 @default.
- W4385488822 creator A5032504499 @default.
- W4385488822 creator A5045895968 @default.
- W4385488822 creator A5084937471 @default.
- W4385488822 date "2023-06-18" @default.
- W4385488822 modified "2023-09-23" @default.
- W4385488822 title "T3S: Improving Multi-Task Reinforcement Learning with Task-Specific Feature Selector and Scheduler" @default.
- W4385488822 cites W2089217417 @default.
- W4385488822 cites W2145339207 @default.
- W4385488822 cites W2166519220 @default.
- W4385488822 cites W2257979135 @default.
- W4385488822 cites W2296073425 @default.
- W4385488822 cites W2809290718 @default.
- W4385488822 cites W2963072899 @default.
- W4385488822 cites W2963430933 @default.
- W4385488822 cites W3000499753 @default.
- W4385488822 cites W3087931390 @default.
- W4385488822 cites W3100944043 @default.
- W4385488822 cites W3121095832 @default.
- W4385488822 cites W3126321819 @default.
- W4385488822 cites W3176624977 @default.
- W4385488822 cites W3205279237 @default.
- W4385488822 doi "https://doi.org/10.1109/ijcnn54540.2023.10191536" @default.
- W4385488822 hasPublicationYear "2023" @default.
- W4385488822 type Work @default.
- W4385488822 citedByCount "0" @default.
- W4385488822 crossrefType "proceedings-article" @default.
- W4385488822 hasAuthorship W4385488822A5013972839 @default.
- W4385488822 hasAuthorship W4385488822A5017719253 @default.
- W4385488822 hasAuthorship W4385488822A5032504499 @default.
- W4385488822 hasAuthorship W4385488822A5045895968 @default.
- W4385488822 hasAuthorship W4385488822A5084937471 @default.
- W4385488822 hasConcept C119857082 @default.
- W4385488822 hasConcept C127413603 @default.
- W4385488822 hasConcept C138885662 @default.
- W4385488822 hasConcept C154945302 @default.
- W4385488822 hasConcept C175154964 @default.
- W4385488822 hasConcept C199360897 @default.
- W4385488822 hasConcept C201995342 @default.
- W4385488822 hasConcept C2776401178 @default.
- W4385488822 hasConcept C2780451532 @default.
- W4385488822 hasConcept C2780801425 @default.
- W4385488822 hasConcept C28006648 @default.
- W4385488822 hasConcept C41008148 @default.
- W4385488822 hasConcept C41895202 @default.
- W4385488822 hasConcept C59404180 @default.
- W4385488822 hasConcept C97541855 @default.
- W4385488822 hasConceptScore W4385488822C119857082 @default.
- W4385488822 hasConceptScore W4385488822C127413603 @default.
- W4385488822 hasConceptScore W4385488822C138885662 @default.
- W4385488822 hasConceptScore W4385488822C154945302 @default.
- W4385488822 hasConceptScore W4385488822C175154964 @default.
- W4385488822 hasConceptScore W4385488822C199360897 @default.
- W4385488822 hasConceptScore W4385488822C201995342 @default.
- W4385488822 hasConceptScore W4385488822C2776401178 @default.
- W4385488822 hasConceptScore W4385488822C2780451532 @default.
- W4385488822 hasConceptScore W4385488822C2780801425 @default.
- W4385488822 hasConceptScore W4385488822C28006648 @default.
- W4385488822 hasConceptScore W4385488822C41008148 @default.
- W4385488822 hasConceptScore W4385488822C41895202 @default.
- W4385488822 hasConceptScore W4385488822C59404180 @default.
- W4385488822 hasConceptScore W4385488822C97541855 @default.
- W4385488822 hasLocation W43854888221 @default.
- W4385488822 hasOpenAccess W4385488822 @default.
- W4385488822 hasPrimaryLocation W43854888221 @default.
- W4385488822 hasRelatedWork W2784094750 @default.
- W4385488822 hasRelatedWork W2895782870 @default.
- W4385488822 hasRelatedWork W2961085424 @default.
- W4385488822 hasRelatedWork W4200580761 @default.
- W4385488822 hasRelatedWork W4292829106 @default.
- W4385488822 hasRelatedWork W4319083788 @default.
- W4385488822 hasRelatedWork W4319309271 @default.
- W4385488822 hasRelatedWork W4366320140 @default.
- W4385488822 hasRelatedWork W4372260038 @default.
- W4385488822 hasRelatedWork W4379662533 @default.
- W4385488822 isParatext "false" @default.
- W4385488822 isRetracted "false" @default.
- W4385488822 workType "article" @default.