Matches in SemOpenAlex for { <https://semopenalex.org/work/W2572448624> ?p ?o ?g. }
- W2572448624 abstract "Deep Reinforcement Learning has enabled the learning of policies for complex tasks in partially observable environments, without explicitly learning the underlying model of the tasks. While such model-free methods achieve considerable performance, they often ignore the structure of task. We present a natural representation of to Reinforcement Learning (RL) problems using Recurrent Convolutional Neural Networks (RCNNs), to better exploit this inherent structure. We define 3 such RCNNs, whose forward passes execute an efficient Value Iteration, propagate beliefs of state in partially observable environments, and choose optimal actions respectively. Backpropagating gradients through these RCNNs allows the system to explicitly learn the Transition Model and Reward Function associated with the underlying MDP, serving as an elegant alternative to classical model-based RL. We evaluate the proposed algorithms in simulation, considering a robot planning problem. We demonstrate the capability of our framework to reduce the cost of replanning, learn accurate MDP models, and finally re-plan with learnt models to achieve near-optimal policies." @default.
- W2572448624 created "2017-01-26" @default.
- W2572448624 creator A5041198151 @default.
- W2572448624 creator A5041308578 @default.
- W2572448624 creator A5063410145 @default.
- W2572448624 date "2017-01-09" @default.
- W2572448624 modified "2023-09-27" @default.
- W2572448624 title "Reinforcement Learning via Recurrent Convolutional Neural Networks" @default.
- W2572448624 cites W1594849649 @default.
- W2572448624 cites W1757796397 @default.
- W2572448624 cites W1977655452 @default.
- W2572448624 cites W1999874108 @default.
- W2572448624 cites W2096533821 @default.
- W2572448624 cites W2115450028 @default.
- W2572448624 cites W2117629901 @default.
- W2572448624 cites W2118688707 @default.
- W2572448624 cites W2141559645 @default.
- W2572448624 cites W2151250975 @default.
- W2572448624 cites W2155027007 @default.
- W2572448624 cites W2258731934 @default.
- W2572448624 cites W2294668865 @default.
- W2572448624 cites W2489939061 @default.
- W2572448624 cites W2952523895 @default.
- W2572448624 cites W2962938178 @default.
- W2572448624 cites W779494576 @default.
- W2572448624 hasPublicationYear "2017" @default.
- W2572448624 type Work @default.
- W2572448624 sameAs 2572448624 @default.
- W2572448624 citedByCount "1" @default.
- W2572448624 countsByYear W25724486242017 @default.
- W2572448624 crossrefType "posted-content" @default.
- W2572448624 hasAuthorship W2572448624A5041198151 @default.
- W2572448624 hasAuthorship W2572448624A5041308578 @default.
- W2572448624 hasAuthorship W2572448624A5063410145 @default.
- W2572448624 hasConcept C119857082 @default.
- W2572448624 hasConcept C121332964 @default.
- W2572448624 hasConcept C126255220 @default.
- W2572448624 hasConcept C14036430 @default.
- W2572448624 hasConcept C14646407 @default.
- W2572448624 hasConcept C154945302 @default.
- W2572448624 hasConcept C162324750 @default.
- W2572448624 hasConcept C165696696 @default.
- W2572448624 hasConcept C17744445 @default.
- W2572448624 hasConcept C187736073 @default.
- W2572448624 hasConcept C199539241 @default.
- W2572448624 hasConcept C2776359362 @default.
- W2572448624 hasConcept C2780451532 @default.
- W2572448624 hasConcept C32848918 @default.
- W2572448624 hasConcept C33923547 @default.
- W2572448624 hasConcept C38652104 @default.
- W2572448624 hasConcept C41008148 @default.
- W2572448624 hasConcept C62520636 @default.
- W2572448624 hasConcept C78458016 @default.
- W2572448624 hasConcept C81363708 @default.
- W2572448624 hasConcept C86803240 @default.
- W2572448624 hasConcept C94625758 @default.
- W2572448624 hasConcept C97541855 @default.
- W2572448624 hasConceptScore W2572448624C119857082 @default.
- W2572448624 hasConceptScore W2572448624C121332964 @default.
- W2572448624 hasConceptScore W2572448624C126255220 @default.
- W2572448624 hasConceptScore W2572448624C14036430 @default.
- W2572448624 hasConceptScore W2572448624C14646407 @default.
- W2572448624 hasConceptScore W2572448624C154945302 @default.
- W2572448624 hasConceptScore W2572448624C162324750 @default.
- W2572448624 hasConceptScore W2572448624C165696696 @default.
- W2572448624 hasConceptScore W2572448624C17744445 @default.
- W2572448624 hasConceptScore W2572448624C187736073 @default.
- W2572448624 hasConceptScore W2572448624C199539241 @default.
- W2572448624 hasConceptScore W2572448624C2776359362 @default.
- W2572448624 hasConceptScore W2572448624C2780451532 @default.
- W2572448624 hasConceptScore W2572448624C32848918 @default.
- W2572448624 hasConceptScore W2572448624C33923547 @default.
- W2572448624 hasConceptScore W2572448624C38652104 @default.
- W2572448624 hasConceptScore W2572448624C41008148 @default.
- W2572448624 hasConceptScore W2572448624C62520636 @default.
- W2572448624 hasConceptScore W2572448624C78458016 @default.
- W2572448624 hasConceptScore W2572448624C81363708 @default.
- W2572448624 hasConceptScore W2572448624C86803240 @default.
- W2572448624 hasConceptScore W2572448624C94625758 @default.
- W2572448624 hasConceptScore W2572448624C97541855 @default.
- W2572448624 hasLocation W25724486241 @default.
- W2572448624 hasOpenAccess W2572448624 @default.
- W2572448624 hasPrimaryLocation W25724486241 @default.
- W2572448624 hasRelatedWork W143164768 @default.
- W2572448624 hasRelatedWork W1552148478 @default.
- W2572448624 hasRelatedWork W1601640269 @default.
- W2572448624 hasRelatedWork W2062122188 @default.
- W2572448624 hasRelatedWork W2097797606 @default.
- W2572448624 hasRelatedWork W2466611570 @default.
- W2572448624 hasRelatedWork W2514775068 @default.
- W2572448624 hasRelatedWork W2515409829 @default.
- W2572448624 hasRelatedWork W2539164647 @default.
- W2572448624 hasRelatedWork W2906650924 @default.
- W2572448624 hasRelatedWork W2915060045 @default.
- W2572448624 hasRelatedWork W2939075711 @default.
- W2572448624 hasRelatedWork W2962854145 @default.
- W2572448624 hasRelatedWork W2963980401 @default.
- W2572448624 hasRelatedWork W2967060478 @default.
- W2572448624 hasRelatedWork W3072315125 @default.
- W2572448624 hasRelatedWork W3139073295 @default.