Matches in SemOpenAlex for { <https://semopenalex.org/work/W3009379771> ?p ?o ?g. }
- W3009379771 abstract "Deep learning in combination with improved training techniques and high computational power has led to recent advances in the field of reinforcement learning (RL) and to successful robotic RL applications such as in-hand manipulation. However, most robotic RL relies on a well known initial state distribution. In real-world tasks, this information is however often not available. For example, when disentangling waste objects the actual position of the robot w.r.t. the objects may not match the positions the RL policy was trained for. To solve this problem, we present a novel adversarial reinforcement learning (ARL) framework. The ARL framework utilizes an adversary, which is trained to steer the original agent, the protagonist, to challenging states. We train the protagonist and the adversary jointly to allow them to adapt to the changing policy of their opponent. We show that our method can generalize from training to test scenarios by training an end-to-end system for robot control to solve a challenging object disentangling task. Experiments with a KUKA LBR+ 7-DOF robot arm show that our approach outperforms the baseline method in disentangling when starting from different initial states than provided during training." @default.
- W3009379771 created "2020-03-13" @default.
- W3009379771 creator A5014519249 @default.
- W3009379771 creator A5017983137 @default.
- W3009379771 creator A5025962742 @default.
- W3009379771 creator A5071367253 @default.
- W3009379771 date "2020-03-08" @default.
- W3009379771 modified "2023-09-27" @default.
- W3009379771 title "Deep Adversarial Reinforcement Learning for Object Disentangling" @default.
- W3009379771 cites W1757796397 @default.
- W3009379771 cites W2047191624 @default.
- W3009379771 cites W2099471712 @default.
- W3009379771 cites W2145339207 @default.
- W3009379771 cites W2257979135 @default.
- W3009379771 cites W2296073425 @default.
- W3009379771 cites W2342840547 @default.
- W3009379771 cites W2601066903 @default.
- W3009379771 cites W2602963933 @default.
- W3009379771 cites W2772709170 @default.
- W3009379771 cites W2773525213 @default.
- W3009379771 cites W2773691349 @default.
- W3009379771 cites W2785962646 @default.
- W3009379771 cites W2908460759 @default.
- W3009379771 cites W2928153079 @default.
- W3009379771 cites W2949103145 @default.
- W3009379771 cites W2963207607 @default.
- W3009379771 cites W2963277051 @default.
- W3009379771 cites W2963293881 @default.
- W3009379771 cites W2963311874 @default.
- W3009379771 cites W2963577640 @default.
- W3009379771 cites W2963684914 @default.
- W3009379771 cites W2981030070 @default.
- W3009379771 cites W2982316857 @default.
- W3009379771 cites W2996037775 @default.
- W3009379771 cites W2996343955 @default.
- W3009379771 cites W3037207827 @default.
- W3009379771 cites W3039584045 @default.
- W3009379771 hasPublicationYear "2020" @default.
- W3009379771 type Work @default.
- W3009379771 sameAs 3009379771 @default.
- W3009379771 citedByCount "0" @default.
- W3009379771 crossrefType "posted-content" @default.
- W3009379771 hasAuthorship W3009379771A5014519249 @default.
- W3009379771 hasAuthorship W3009379771A5017983137 @default.
- W3009379771 hasAuthorship W3009379771A5025962742 @default.
- W3009379771 hasAuthorship W3009379771A5071367253 @default.
- W3009379771 hasConcept C10138342 @default.
- W3009379771 hasConcept C119857082 @default.
- W3009379771 hasConcept C127413603 @default.
- W3009379771 hasConcept C150415221 @default.
- W3009379771 hasConcept C154945302 @default.
- W3009379771 hasConcept C162324750 @default.
- W3009379771 hasConcept C198082294 @default.
- W3009379771 hasConcept C201995342 @default.
- W3009379771 hasConcept C2775924081 @default.
- W3009379771 hasConcept C2780451532 @default.
- W3009379771 hasConcept C2781238097 @default.
- W3009379771 hasConcept C34413123 @default.
- W3009379771 hasConcept C37736160 @default.
- W3009379771 hasConcept C38652104 @default.
- W3009379771 hasConcept C41008148 @default.
- W3009379771 hasConcept C41065033 @default.
- W3009379771 hasConcept C90509273 @default.
- W3009379771 hasConcept C97541855 @default.
- W3009379771 hasConceptScore W3009379771C10138342 @default.
- W3009379771 hasConceptScore W3009379771C119857082 @default.
- W3009379771 hasConceptScore W3009379771C127413603 @default.
- W3009379771 hasConceptScore W3009379771C150415221 @default.
- W3009379771 hasConceptScore W3009379771C154945302 @default.
- W3009379771 hasConceptScore W3009379771C162324750 @default.
- W3009379771 hasConceptScore W3009379771C198082294 @default.
- W3009379771 hasConceptScore W3009379771C201995342 @default.
- W3009379771 hasConceptScore W3009379771C2775924081 @default.
- W3009379771 hasConceptScore W3009379771C2780451532 @default.
- W3009379771 hasConceptScore W3009379771C2781238097 @default.
- W3009379771 hasConceptScore W3009379771C34413123 @default.
- W3009379771 hasConceptScore W3009379771C37736160 @default.
- W3009379771 hasConceptScore W3009379771C38652104 @default.
- W3009379771 hasConceptScore W3009379771C41008148 @default.
- W3009379771 hasConceptScore W3009379771C41065033 @default.
- W3009379771 hasConceptScore W3009379771C90509273 @default.
- W3009379771 hasConceptScore W3009379771C97541855 @default.
- W3009379771 hasLocation W30093797711 @default.
- W3009379771 hasOpenAccess W3009379771 @default.
- W3009379771 hasPrimaryLocation W30093797711 @default.
- W3009379771 hasRelatedWork W1972063518 @default.
- W3009379771 hasRelatedWork W2513373085 @default.
- W3009379771 hasRelatedWork W2528734395 @default.
- W3009379771 hasRelatedWork W2733961795 @default.
- W3009379771 hasRelatedWork W2735268712 @default.
- W3009379771 hasRelatedWork W2889878992 @default.
- W3009379771 hasRelatedWork W2910219310 @default.
- W3009379771 hasRelatedWork W2914584948 @default.
- W3009379771 hasRelatedWork W2947525798 @default.
- W3009379771 hasRelatedWork W2952672470 @default.
- W3009379771 hasRelatedWork W2953129341 @default.
- W3009379771 hasRelatedWork W2964198579 @default.
- W3009379771 hasRelatedWork W2973387504 @default.
- W3009379771 hasRelatedWork W3035756007 @default.
- W3009379771 hasRelatedWork W3133374884 @default.