Matches in SemOpenAlex for { <https://semopenalex.org/work/W2757789407> ?p ?o ?g. }
- W2757789407 abstract "Deep reinforcement learning yields great results for a large array of problems, but models are generally retrained anew for each new problem to be solved. Prior learning and knowledge are difficult to incorporate when training new models, requiring increasingly longer training as problems become more complex. This is especially problematic for problems with sparse rewards. We provide a solution to these problems by introducing Concept Network Reinforcement Learning (CNRL), a framework which allows us to decompose problems using a multi-level hierarchy. Concepts in a concept network are reusable, and flexible enough to encapsulate feature extractors, skills, or other concept networks. With this hierarchical learning approach, deep reinforcement learning can be used to solve complex tasks in a modular way, through problem decomposition. We demonstrate the strength of CNRL by training a model to grasp a rectangular prism and precisely stack it on top of a cube using a gripper on a Kinova JACO arm, simulated in MuJoCo. Our experiments show that our use of hierarchy results in a 45x reduction in environment interactions compared to the state-of-the-art on this task." @default.
- W2757789407 created "2017-10-06" @default.
- W2757789407 creator A5001271123 @default.
- W2757789407 creator A5008727361 @default.
- W2757789407 creator A5037120683 @default.
- W2757789407 creator A5046717011 @default.
- W2757789407 creator A5068542686 @default.
- W2757789407 creator A5078322249 @default.
- W2757789407 creator A5084533122 @default.
- W2757789407 date "2017-09-20" @default.
- W2757789407 modified "2023-09-26" @default.
- W2757789407 title "Deep Reinforcement Learning for Dexterous Manipulation with Concept Networks." @default.
- W2757789407 cites W1191599655 @default.
- W2757789407 cites W1575592356 @default.
- W2757789407 cites W1585861384 @default.
- W2757789407 cites W1771410628 @default.
- W2757789407 cites W1994923984 @default.
- W2757789407 cites W2064076655 @default.
- W2757789407 cites W2101677491 @default.
- W2757789407 cites W2109910161 @default.
- W2757789407 cites W2145339207 @default.
- W2757789407 cites W2163302320 @default.
- W2757789407 cites W2165150801 @default.
- W2757789407 cites W2173248099 @default.
- W2757789407 cites W2257979135 @default.
- W2757789407 cites W2293467699 @default.
- W2757789407 cites W2335959470 @default.
- W2757789407 cites W2344023930 @default.
- W2757789407 cites W2610395436 @default.
- W2757789407 cites W2953326790 @default.
- W2757789407 cites W2963590100 @default.
- W2757789407 hasPublicationYear "2017" @default.
- W2757789407 type Work @default.
- W2757789407 sameAs 2757789407 @default.
- W2757789407 citedByCount "8" @default.
- W2757789407 countsByYear W27577894072018 @default.
- W2757789407 countsByYear W27577894072019 @default.
- W2757789407 countsByYear W27577894072020 @default.
- W2757789407 countsByYear W27577894072023 @default.
- W2757789407 crossrefType "posted-content" @default.
- W2757789407 hasAuthorship W2757789407A5001271123 @default.
- W2757789407 hasAuthorship W2757789407A5008727361 @default.
- W2757789407 hasAuthorship W2757789407A5037120683 @default.
- W2757789407 hasAuthorship W2757789407A5046717011 @default.
- W2757789407 hasAuthorship W2757789407A5068542686 @default.
- W2757789407 hasAuthorship W2757789407A5078322249 @default.
- W2757789407 hasAuthorship W2757789407A5084533122 @default.
- W2757789407 hasConcept C101468663 @default.
- W2757789407 hasConcept C111335779 @default.
- W2757789407 hasConcept C111919701 @default.
- W2757789407 hasConcept C115903868 @default.
- W2757789407 hasConcept C119857082 @default.
- W2757789407 hasConcept C124681953 @default.
- W2757789407 hasConcept C127413603 @default.
- W2757789407 hasConcept C138885662 @default.
- W2757789407 hasConcept C154945302 @default.
- W2757789407 hasConcept C162324750 @default.
- W2757789407 hasConcept C171268870 @default.
- W2757789407 hasConcept C18903297 @default.
- W2757789407 hasConcept C201995342 @default.
- W2757789407 hasConcept C2524010 @default.
- W2757789407 hasConcept C2776401178 @default.
- W2757789407 hasConcept C2780451532 @default.
- W2757789407 hasConcept C31170391 @default.
- W2757789407 hasConcept C33923547 @default.
- W2757789407 hasConcept C34447519 @default.
- W2757789407 hasConcept C41008148 @default.
- W2757789407 hasConcept C41895202 @default.
- W2757789407 hasConcept C86803240 @default.
- W2757789407 hasConcept C97541855 @default.
- W2757789407 hasConceptScore W2757789407C101468663 @default.
- W2757789407 hasConceptScore W2757789407C111335779 @default.
- W2757789407 hasConceptScore W2757789407C111919701 @default.
- W2757789407 hasConceptScore W2757789407C115903868 @default.
- W2757789407 hasConceptScore W2757789407C119857082 @default.
- W2757789407 hasConceptScore W2757789407C124681953 @default.
- W2757789407 hasConceptScore W2757789407C127413603 @default.
- W2757789407 hasConceptScore W2757789407C138885662 @default.
- W2757789407 hasConceptScore W2757789407C154945302 @default.
- W2757789407 hasConceptScore W2757789407C162324750 @default.
- W2757789407 hasConceptScore W2757789407C171268870 @default.
- W2757789407 hasConceptScore W2757789407C18903297 @default.
- W2757789407 hasConceptScore W2757789407C201995342 @default.
- W2757789407 hasConceptScore W2757789407C2524010 @default.
- W2757789407 hasConceptScore W2757789407C2776401178 @default.
- W2757789407 hasConceptScore W2757789407C2780451532 @default.
- W2757789407 hasConceptScore W2757789407C31170391 @default.
- W2757789407 hasConceptScore W2757789407C33923547 @default.
- W2757789407 hasConceptScore W2757789407C34447519 @default.
- W2757789407 hasConceptScore W2757789407C41008148 @default.
- W2757789407 hasConceptScore W2757789407C41895202 @default.
- W2757789407 hasConceptScore W2757789407C86803240 @default.
- W2757789407 hasConceptScore W2757789407C97541855 @default.
- W2757789407 hasLocation W27577894071 @default.
- W2757789407 hasOpenAccess W2757789407 @default.
- W2757789407 hasPrimaryLocation W27577894071 @default.
- W2757789407 hasRelatedWork W1555368087 @default.
- W2757789407 hasRelatedWork W1595483645 @default.
- W2757789407 hasRelatedWork W2158150115 @default.
- W2757789407 hasRelatedWork W2304525083 @default.