FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2943686261> ?p ?o ?g. }

• W2943686261 abstract "Recurrent neural networks (RNNs) for reinforcement learning (RL) have shown distinct advantages, e.g., solving memory-dependent tasks and meta-learning. However, little effort has been spent on improving RNN architectures and on understanding the underlying neural mechanisms for performance gain. In this paper, we propose a novel, multiple-timescale, stochastic RNN for RL. Empirical results show that the network can autonomously learn to abstract sub-goals and can self-develop an action hierarchy using internal dynamics in a challenging continuous control task. Furthermore, we show that the self-developed compositionality of the network enhances faster re-learning when adapting to a new task that is a re-composition of previously learned sub-goals, than when starting from scratch. We also found that improved performance can be achieved when neural activities are subject to stochastic rather than deterministic dynamics." @default.
• W2943686261 created "2019-05-09" @default.
• W2943686261 creator A5000727773 @default.
• W2943686261 creator A5004840638 @default.
• W2943686261 creator A5030709757 @default.
• W2943686261 date "2019-01-29" @default.
• W2943686261 modified "2023-09-23" @default.
• W2943686261 title "Self-organization of action hierarchy and compositionality by reinforcement learning with recurrent neural networks" @default.
• W2943686261 cites W1506440042 @default.
• W2943686261 cites W1542791059 @default.
• W2943686261 cites W1652352837 @default.
• W2943686261 cites W1662842982 @default.
• W2943686261 cites W1846027912 @default.
• W2943686261 cites W1980006923 @default.
• W2943686261 cites W2003565109 @default.
• W2943686261 cites W2011418219 @default.
• W2943686261 cites W2017957151 @default.
• W2943686261 cites W2028418738 @default.
• W2943686261 cites W2028612696 @default.
• W2943686261 cites W2032610016 @default.
• W2943686261 cites W2064675550 @default.
• W2943686261 cites W2066835186 @default.
• W2943686261 cites W2076337359 @default.
• W2943686261 cites W2091026426 @default.
• W2943686261 cites W2100259206 @default.
• W2943686261 cites W2109910161 @default.
• W2943686261 cites W2113501460 @default.
• W2943686261 cites W2115121720 @default.
• W2943686261 cites W2120501001 @default.
• W2943686261 cites W2121517924 @default.
• W2943686261 cites W2133458291 @default.
• W2943686261 cites W2135353545 @default.
• W2943686261 cites W2140135625 @default.
• W2943686261 cites W2144984120 @default.
• W2943686261 cites W2145339207 @default.
• W2943686261 cites W2149112028 @default.
• W2943686261 cites W2150355110 @default.
• W2943686261 cites W2155027007 @default.
• W2943686261 cites W2160644528 @default.
• W2943686261 cites W2172294131 @default.
• W2943686261 cites W2173248099 @default.
• W2943686261 cites W2203645894 @default.
• W2943686261 cites W2257979135 @default.
• W2943686261 cites W2291973609 @default.
• W2943686261 cites W2335959470 @default.
• W2943686261 cites W2472819217 @default.
• W2943686261 cites W2504933562 @default.
• W2943686261 cites W2538677090 @default.
• W2943686261 cites W2550182557 @default.
• W2943686261 cites W2605016475 @default.
• W2943686261 cites W2736601468 @default.
• W2943686261 cites W2737989178 @default.
• W2943686261 cites W2761873684 @default.
• W2943686261 cites W2762872434 @default.
• W2943686261 cites W2766447205 @default.
• W2943686261 cites W2768760676 @default.
• W2943686261 cites W2781726626 @default.
• W2943686261 cites W2788385947 @default.
• W2943686261 cites W2805481182 @default.
• W2943686261 cites W2887671224 @default.
• W2943686261 cites W2890208753 @default.
• W2943686261 cites W2908064123 @default.
• W2943686261 cites W2908124316 @default.
• W2943686261 cites W2920362155 @default.
• W2943686261 cites W2949267040 @default.
• W2943686261 cites W2949369413 @default.
• W2943686261 cites W2951004968 @default.
• W2943686261 cites W2951775809 @default.
• W2943686261 cites W2953334758 @default.
• W2943686261 cites W2962938178 @default.
• W2943686261 cites W2963095800 @default.
• W2943686261 cites W2963161674 @default.
• W2943686261 cites W2963279312 @default.
• W2943686261 cites W2963286043 @default.
• W2943686261 cites W2963454359 @default.
• W2943686261 cites W2963912551 @default.
• W2943686261 cites W2964043796 @default.
• W2943686261 cites W2964078140 @default.
• W2943686261 cites W2964118262 @default.
• W2943686261 cites W2964121937 @default.
• W2943686261 cites W2964174623 @default.
• W2943686261 cites W2964227312 @default.
• W2943686261 cites W2974745023 @default.
• W2943686261 cites W3033314805 @default.
• W2943686261 cites W592244745 @default.
• W2943686261 cites W2426267443 @default.
• W2943686261 doi "https://doi.org/10.48550/arxiv.1901.10113" @default.
• W2943686261 hasPublicationYear "2019" @default.
• W2943686261 type Work @default.
• W2943686261 sameAs 2943686261 @default.
• W2943686261 citedByCount "4" @default.
• W2943686261 countsByYear W29436862612019 @default.
• W2943686261 countsByYear W29436862612020 @default.
• W2943686261 countsByYear W29436862612021 @default.
• W2943686261 crossrefType "posted-content" @default.
• W2943686261 hasAuthorship W2943686261A5000727773 @default.
• W2943686261 hasAuthorship W2943686261A5004840638 @default.
• W2943686261 hasAuthorship W2943686261A5030709757 @default.
• W2943686261 hasBestOaLocation W29436862611 @default.
• W2943686261 hasConcept C119857082 @default.

• next