Matches in SemOpenAlex for { <https://semopenalex.org/work/W3023392004> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W3023392004 abstract "Planning is a powerful approach to control problems with known environment dynamics. In unknown environments the agent needs to learn a model of the system dynamics to make planning applicable. This is particularly challenging when the underlying states are only indirectly observable through images. We propose to learn a deep latent Gaussian process dynamics (DLGPD) model that learns low-dimensional system dynamics from environment interactions with visual observations. The method infers latent state representations from observations using neural networks and models the system dynamics in the learned latent space with Gaussian processes. All parts of the model can be trained jointly by optimizing a lower bound on the likelihood of transitions in image space. We evaluate the proposed approach on the pendulum swing-up task while using the learned dynamics model for planning in latent space in order to solve the control problem. We also demonstrate that our method can quickly adapt a trained agent to changes in the system dynamics from just a few rollouts. We compare our approach to a state-of-the-art purely deep learning based method and demonstrate the advantages of combining Gaussian processes with deep learning for data efficiency and transfer learning." @default.
- W3023392004 created "2020-05-13" @default.
- W3023392004 creator A5019500529 @default.
- W3023392004 creator A5042165162 @default.
- W3023392004 creator A5085424314 @default.
- W3023392004 creator A5090283032 @default.
- W3023392004 date "2020-05-07" @default.
- W3023392004 modified "2023-09-25" @default.
- W3023392004 title "Planning from Images with Deep Latent Gaussian Process Dynamics" @default.
- W3023392004 cites W1522301498 @default.
- W3023392004 cites W164706946 @default.
- W3023392004 cites W1959608418 @default.
- W3023392004 cites W1985658808 @default.
- W3023392004 cites W2073787051 @default.
- W3023392004 cites W2132083787 @default.
- W3023392004 cites W2140135625 @default.
- W3023392004 cites W2145339207 @default.
- W3023392004 cites W2151268438 @default.
- W3023392004 cites W2154032554 @default.
- W3023392004 cites W2162717641 @default.
- W3023392004 cites W2166063021 @default.
- W3023392004 cites W2181909627 @default.
- W3023392004 cites W2251955814 @default.
- W3023392004 cites W2418628973 @default.
- W3023392004 cites W2575705757 @default.
- W3023392004 cites W2787666871 @default.
- W3023392004 cites W2809717747 @default.
- W3023392004 cites W2889347284 @default.
- W3023392004 cites W2893995718 @default.
- W3023392004 cites W2900152462 @default.
- W3023392004 cites W2911894378 @default.
- W3023392004 cites W2962897886 @default.
- W3023392004 cites W2963166838 @default.
- W3023392004 cites W2963190151 @default.
- W3023392004 cites W2963402509 @default.
- W3023392004 cites W2963412383 @default.
- W3023392004 cites W2963430173 @default.
- W3023392004 cites W2963864421 @default.
- W3023392004 cites W2963887494 @default.
- W3023392004 cites W2964267629 @default.
- W3023392004 cites W3093010610 @default.
- W3023392004 cites W391985582 @default.
- W3023392004 hasPublicationYear "2020" @default.
- W3023392004 type Work @default.
- W3023392004 sameAs 3023392004 @default.
- W3023392004 citedByCount "0" @default.
- W3023392004 crossrefType "posted-content" @default.
- W3023392004 hasAuthorship W3023392004A5019500529 @default.
- W3023392004 hasAuthorship W3023392004A5042165162 @default.
- W3023392004 hasAuthorship W3023392004A5085424314 @default.
- W3023392004 hasAuthorship W3023392004A5090283032 @default.
- W3023392004 hasConcept C108583219 @default.
- W3023392004 hasConcept C111919701 @default.
- W3023392004 hasConcept C119857082 @default.
- W3023392004 hasConcept C121332964 @default.
- W3023392004 hasConcept C154945302 @default.
- W3023392004 hasConcept C163716315 @default.
- W3023392004 hasConcept C41008148 @default.
- W3023392004 hasConcept C61326573 @default.
- W3023392004 hasConcept C62520636 @default.
- W3023392004 hasConcept C77405623 @default.
- W3023392004 hasConcept C98045186 @default.
- W3023392004 hasConceptScore W3023392004C108583219 @default.
- W3023392004 hasConceptScore W3023392004C111919701 @default.
- W3023392004 hasConceptScore W3023392004C119857082 @default.
- W3023392004 hasConceptScore W3023392004C121332964 @default.
- W3023392004 hasConceptScore W3023392004C154945302 @default.
- W3023392004 hasConceptScore W3023392004C163716315 @default.
- W3023392004 hasConceptScore W3023392004C41008148 @default.
- W3023392004 hasConceptScore W3023392004C61326573 @default.
- W3023392004 hasConceptScore W3023392004C62520636 @default.
- W3023392004 hasConceptScore W3023392004C77405623 @default.
- W3023392004 hasConceptScore W3023392004C98045186 @default.
- W3023392004 hasOpenAccess W3023392004 @default.
- W3023392004 hasRelatedWork W16304275 @default.
- W3023392004 hasRelatedWork W2939673055 @default.
- W3023392004 hasRelatedWork W2949477558 @default.
- W3023392004 hasRelatedWork W2951021014 @default.
- W3023392004 hasRelatedWork W2951748364 @default.
- W3023392004 hasRelatedWork W2963166838 @default.
- W3023392004 hasRelatedWork W2980073099 @default.
- W3023392004 hasRelatedWork W2982437637 @default.
- W3023392004 hasRelatedWork W2987301161 @default.
- W3023392004 hasRelatedWork W2994512474 @default.
- W3023392004 hasRelatedWork W3024287798 @default.
- W3023392004 hasRelatedWork W3085647390 @default.
- W3023392004 hasRelatedWork W3098168339 @default.
- W3023392004 hasRelatedWork W3102488324 @default.
- W3023392004 hasRelatedWork W3104725582 @default.
- W3023392004 hasRelatedWork W3126417600 @default.
- W3023392004 hasRelatedWork W3166705655 @default.
- W3023392004 hasRelatedWork W3198861570 @default.
- W3023392004 hasRelatedWork W3207748996 @default.
- W3023392004 hasRelatedWork W3210549225 @default.
- W3023392004 isParatext "false" @default.
- W3023392004 isRetracted "false" @default.
- W3023392004 magId "3023392004" @default.
- W3023392004 workType "article" @default.