FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048943668> ?p ?o ?g. }

• W3048943668 abstract "Deep Reinforcement Learning (RL) has shown great success in learning complex control policies for a variety of applications in robotics. However, in most such cases, the hardware of the robot has been considered immutable, modeled as part of the environment. In this study, we explore the problem of learning hardware and control parameters together in a unified RL framework. To achieve this, we propose to model aspects of the robot's hardware as a policy, analogous to and optimized jointly with its computational counterpart. We show that, by modeling such mechanical policies as auto-differentiable computational graphs, the ensuing optimization problem can be solved efficiently by gradient-based algorithms from the Policy Optimization family. We present two such design examples: a toy mass-spring problem, and a real-world problem of designing an underactuated hand. We compare our method against traditional co-optimization approaches, and also demonstrate its effectiveness by building a physical prototype based on the learned hardware parameters." @default.
• W3048943668 created "2020-08-18" @default.
• W3048943668 creator A5036043911 @default.
• W3048943668 creator A5068882233 @default.
• W3048943668 creator A5087490971 @default.
• W3048943668 date "2020-08-11" @default.
• W3048943668 modified "2023-09-27" @default.
• W3048943668 title "Hardware as Policy: Mechanical and Computational Co-Optimization using Deep Reinforcement Learning" @default.
• W3048943668 cites W1609223620 @default.
• W3048943668 cites W1771410628 @default.
• W3048943668 cites W1980928215 @default.
• W3048943668 cites W2048058874 @default.
• W3048943668 cites W2093373149 @default.
• W3048943668 cites W2097859469 @default.
• W3048943668 cites W2112036188 @default.
• W3048943668 cites W2117085697 @default.
• W3048943668 cites W2117197536 @default.
• W3048943668 cites W2140190473 @default.
• W3048943668 cites W2158782408 @default.
• W3048943668 cites W2166213162 @default.
• W3048943668 cites W2168661995 @default.
• W3048943668 cites W2173248099 @default.
• W3048943668 cites W2321457851 @default.
• W3048943668 cites W2556096037 @default.
• W3048943668 cites W2605102758 @default.
• W3048943668 cites W2610992585 @default.
• W3048943668 cites W2736601468 @default.
• W3048943668 cites W2757631751 @default.
• W3048943668 cites W2775103860 @default.
• W3048943668 cites W2807229064 @default.
• W3048943668 cites W2885163910 @default.
• W3048943668 cites W2891122218 @default.
• W3048943668 cites W2907537824 @default.
• W3048943668 cites W2949608212 @default.
• W3048943668 cites W2962696179 @default.
• W3048943668 cites W2963411833 @default.
• W3048943668 cites W2963534270 @default.
• W3048943668 cites W2967894826 @default.
• W3048943668 cites W2968042644 @default.
• W3048943668 cites W2978940263 @default.
• W3048943668 cites W3011782672 @default.
• W3048943668 cites W3029831493 @default.
• W3048943668 cites W605157186 @default.
• W3048943668 hasPublicationYear "2020" @default.
• W3048943668 type Work @default.
• W3048943668 sameAs 3048943668 @default.
• W3048943668 citedByCount "7" @default.
• W3048943668 countsByYear W30489436682020 @default.
• W3048943668 countsByYear W30489436682021 @default.
• W3048943668 countsByYear W30489436682022 @default.
• W3048943668 crossrefType "posted-content" @default.
• W3048943668 hasAuthorship W3048943668A5036043911 @default.
• W3048943668 hasAuthorship W3048943668A5068882233 @default.
• W3048943668 hasAuthorship W3048943668A5087490971 @default.
• W3048943668 hasConcept C11413529 @default.
• W3048943668 hasConcept C126255220 @default.
• W3048943668 hasConcept C127413603 @default.
• W3048943668 hasConcept C133731056 @default.
• W3048943668 hasConcept C134306372 @default.
• W3048943668 hasConcept C136197465 @default.
• W3048943668 hasConcept C137836250 @default.
• W3048943668 hasConcept C154945302 @default.
• W3048943668 hasConcept C202615002 @default.
• W3048943668 hasConcept C2775924081 @default.
• W3048943668 hasConcept C33923547 @default.
• W3048943668 hasConcept C34413123 @default.
• W3048943668 hasConcept C41008148 @default.
• W3048943668 hasConcept C88337583 @default.
• W3048943668 hasConcept C90509273 @default.
• W3048943668 hasConcept C97541855 @default.
• W3048943668 hasConceptScore W3048943668C11413529 @default.
• W3048943668 hasConceptScore W3048943668C126255220 @default.
• W3048943668 hasConceptScore W3048943668C127413603 @default.
• W3048943668 hasConceptScore W3048943668C133731056 @default.
• W3048943668 hasConceptScore W3048943668C134306372 @default.
• W3048943668 hasConceptScore W3048943668C136197465 @default.
• W3048943668 hasConceptScore W3048943668C137836250 @default.
• W3048943668 hasConceptScore W3048943668C154945302 @default.
• W3048943668 hasConceptScore W3048943668C202615002 @default.
• W3048943668 hasConceptScore W3048943668C2775924081 @default.
• W3048943668 hasConceptScore W3048943668C33923547 @default.
• W3048943668 hasConceptScore W3048943668C34413123 @default.
• W3048943668 hasConceptScore W3048943668C41008148 @default.
• W3048943668 hasConceptScore W3048943668C88337583 @default.
• W3048943668 hasConceptScore W3048943668C90509273 @default.
• W3048943668 hasConceptScore W3048943668C97541855 @default.
• W3048943668 hasLocation W30489436681 @default.
• W3048943668 hasOpenAccess W3048943668 @default.
• W3048943668 hasPrimaryLocation W30489436681 @default.
• W3048943668 hasRelatedWork W2083703071 @default.
• W3048943668 hasRelatedWork W2092184204 @default.
• W3048943668 hasRelatedWork W2129444665 @default.
• W3048943668 hasRelatedWork W2554684686 @default.
• W3048943668 hasRelatedWork W2604726708 @default.
• W3048943668 hasRelatedWork W2736021157 @default.
• W3048943668 hasRelatedWork W2739930441 @default.
• W3048943668 hasRelatedWork W2909770505 @default.
• W3048943668 hasRelatedWork W2929842775 @default.
• W3048943668 hasRelatedWork W2962696179 @default.
• W3048943668 hasRelatedWork W2970292059 @default.

• next