Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386024530> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W4386024530 endingPage "117" @default.
- W4386024530 startingPage "104" @default.
- W4386024530 abstract "The implementation of robotic reinforcement learning is hampered by problems such as an unspecified reward function and high training costs. Many previous works have used cross-domain policy transfer to obtain the policy of the problem domain. However, these researches require paired and aligned dynamics trajectories or other interactions with the environment. We propose a cross-domain dynamics alignment framework for the problem domain policy acquisition that can transfer the policy trained in the source domain to the problem domain. Our framework aims to learn dynamics alignment across two domains that differ in agents' physical parameters (armature, rotation range, or torso mass) or agents' morphologies (limbs). Most importantly, we learn dynamics alignment between two domains using unpaired and unaligned dynamics trajectories. For these two scenarios, we propose a cross-physics-domain policy adaptation algorithm (CPD) and a cross-morphology-domain policy adaptation algorithm (CMD) based on our cross-domain dynamics alignment framework. In order to improve the performance of policy in the source domain so that a better policy can be transferred to the problem domain, we propose the Boltzmann TD3 (BTD3) algorithm. We conduct diverse experiments on agent continuous control domains to demonstrate the performance of our approaches. Experimental results show that our approaches can obtain better policies and higher rewards for the agents in the problem domains even when the dataset of the problem domain is small." @default.
- W4386024530 created "2023-08-22" @default.
- W4386024530 creator A5002516992 @default.
- W4386024530 creator A5015102287 @default.
- W4386024530 creator A5023507049 @default.
- W4386024530 creator A5054060458 @default.
- W4386024530 creator A5059249318 @default.
- W4386024530 creator A5062045805 @default.
- W4386024530 creator A5080195283 @default.
- W4386024530 creator A5089907846 @default.
- W4386024530 date "2023-10-01" @default.
- W4386024530 modified "2023-10-18" @default.
- W4386024530 title "Cross-domain policy adaptation with dynamics alignment" @default.
- W4386024530 cites W2014512216 @default.
- W4386024530 cites W2051228319 @default.
- W4386024530 cites W3080884797 @default.
- W4386024530 cites W3096831136 @default.
- W4386024530 cites W3195641273 @default.
- W4386024530 cites W4226143977 @default.
- W4386024530 cites W4283769870 @default.
- W4386024530 cites W4292862358 @default.
- W4386024530 cites W4297121106 @default.
- W4386024530 cites W4308784435 @default.
- W4386024530 cites W4313398169 @default.
- W4386024530 cites W4317935412 @default.
- W4386024530 cites W4321457209 @default.
- W4386024530 doi "https://doi.org/10.1016/j.neunet.2023.08.025" @default.
- W4386024530 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37647740" @default.
- W4386024530 hasPublicationYear "2023" @default.
- W4386024530 type Work @default.
- W4386024530 citedByCount "0" @default.
- W4386024530 crossrefType "journal-article" @default.
- W4386024530 hasAuthorship W4386024530A5002516992 @default.
- W4386024530 hasAuthorship W4386024530A5015102287 @default.
- W4386024530 hasAuthorship W4386024530A5023507049 @default.
- W4386024530 hasAuthorship W4386024530A5054060458 @default.
- W4386024530 hasAuthorship W4386024530A5059249318 @default.
- W4386024530 hasAuthorship W4386024530A5062045805 @default.
- W4386024530 hasAuthorship W4386024530A5080195283 @default.
- W4386024530 hasAuthorship W4386024530A5089907846 @default.
- W4386024530 hasConcept C134306372 @default.
- W4386024530 hasConcept C154945302 @default.
- W4386024530 hasConcept C33923547 @default.
- W4386024530 hasConcept C36503486 @default.
- W4386024530 hasConcept C41008148 @default.
- W4386024530 hasConcept C97541855 @default.
- W4386024530 hasConceptScore W4386024530C134306372 @default.
- W4386024530 hasConceptScore W4386024530C154945302 @default.
- W4386024530 hasConceptScore W4386024530C33923547 @default.
- W4386024530 hasConceptScore W4386024530C36503486 @default.
- W4386024530 hasConceptScore W4386024530C41008148 @default.
- W4386024530 hasConceptScore W4386024530C97541855 @default.
- W4386024530 hasLocation W43860245301 @default.
- W4386024530 hasLocation W43860245302 @default.
- W4386024530 hasOpenAccess W4386024530 @default.
- W4386024530 hasPrimaryLocation W43860245301 @default.
- W4386024530 hasRelatedWork W2031695474 @default.
- W4386024530 hasRelatedWork W2138720691 @default.
- W4386024530 hasRelatedWork W2358668433 @default.
- W4386024530 hasRelatedWork W2376932109 @default.
- W4386024530 hasRelatedWork W2390279801 @default.
- W4386024530 hasRelatedWork W2748952813 @default.
- W4386024530 hasRelatedWork W2899084033 @default.
- W4386024530 hasRelatedWork W4306904969 @default.
- W4386024530 hasRelatedWork W4362501864 @default.
- W4386024530 hasRelatedWork W4380318855 @default.
- W4386024530 hasVolume "167" @default.
- W4386024530 isParatext "false" @default.
- W4386024530 isRetracted "false" @default.
- W4386024530 workType "article" @default.