Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285294350> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W4285294350 endingPage "2769" @default.
- W4285294350 startingPage "2759" @default.
- W4285294350 abstract "In multigoal reinforcement learning (RL), algorithms usually suffer from inefficiency in the collection of successful experiences in tasks with sparse rewards. By utilizing the ideas of relabeling hindsight experience and curriculum learning, some prior works have greatly improved the sample efficiency in robotic manipulation tasks, such as hindsight experience replay (HER), hindsight goal generation (HGG), graph-based HGG (G-HGG), and curriculum-guided HER (CHER). However, none of these can learn efficiently to solve challenging manipulation tasks with distant goals and obstacles, since they rely either on heuristic or simple distance-guided exploration. In this article, we introduce graph-curriculum-guided HGG (GC-HGG), an extension of CHER and G-HGG, which works by selecting hindsight goals on the basis of graph-based proximity and diversity. We evaluated GC-HGG in four challenging manipulation tasks involving obstacles in both simulations and real-world experiments, in which significant enhancements in both sample efficiency and overall success rates over prior works were demonstrated. Videos and codes can be viewed at this link: <uri xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>https://videoviewsite.wixsite.com/gc-hgg</uri> ." @default.
- W4285294350 created "2022-07-14" @default.
- W4285294350 creator A5011780593 @default.
- W4285294350 creator A5013497581 @default.
- W4285294350 creator A5018299384 @default.
- W4285294350 creator A5033792439 @default.
- W4285294350 creator A5053561890 @default.
- W4285294350 creator A5060444894 @default.
- W4285294350 creator A5088159741 @default.
- W4285294350 date "2023-03-01" @default.
- W4285294350 modified "2023-09-26" @default.
- W4285294350 title "Solving Robotic Manipulation With Sparse Reward Reinforcement Learning Via Graph-Based Diversity and Proximity" @default.
- W4285294350 cites W2139612737 @default.
- W4285294350 cites W2169528473 @default.
- W4285294350 cites W2963523627 @default.
- W4285294350 cites W2978938326 @default.
- W4285294350 cites W3100789280 @default.
- W4285294350 cites W3109467707 @default.
- W4285294350 cites W3117515147 @default.
- W4285294350 cites W3174508330 @default.
- W4285294350 cites W3175584021 @default.
- W4285294350 cites W3179972165 @default.
- W4285294350 cites W3209777177 @default.
- W4285294350 doi "https://doi.org/10.1109/tie.2022.3172754" @default.
- W4285294350 hasPublicationYear "2023" @default.
- W4285294350 type Work @default.
- W4285294350 citedByCount "2" @default.
- W4285294350 countsByYear W42852943502022 @default.
- W4285294350 countsByYear W42852943502023 @default.
- W4285294350 crossrefType "journal-article" @default.
- W4285294350 hasAuthorship W4285294350A5011780593 @default.
- W4285294350 hasAuthorship W4285294350A5013497581 @default.
- W4285294350 hasAuthorship W4285294350A5018299384 @default.
- W4285294350 hasAuthorship W4285294350A5033792439 @default.
- W4285294350 hasAuthorship W4285294350A5053561890 @default.
- W4285294350 hasAuthorship W4285294350A5060444894 @default.
- W4285294350 hasAuthorship W4285294350A5088159741 @default.
- W4285294350 hasConcept C10347200 @default.
- W4285294350 hasConcept C119857082 @default.
- W4285294350 hasConcept C132525143 @default.
- W4285294350 hasConcept C154945302 @default.
- W4285294350 hasConcept C15744967 @default.
- W4285294350 hasConcept C162324750 @default.
- W4285294350 hasConcept C173801870 @default.
- W4285294350 hasConcept C175444787 @default.
- W4285294350 hasConcept C180747234 @default.
- W4285294350 hasConcept C19417346 @default.
- W4285294350 hasConcept C2778869765 @default.
- W4285294350 hasConcept C41008148 @default.
- W4285294350 hasConcept C47177190 @default.
- W4285294350 hasConcept C80444323 @default.
- W4285294350 hasConcept C97541855 @default.
- W4285294350 hasConceptScore W4285294350C10347200 @default.
- W4285294350 hasConceptScore W4285294350C119857082 @default.
- W4285294350 hasConceptScore W4285294350C132525143 @default.
- W4285294350 hasConceptScore W4285294350C154945302 @default.
- W4285294350 hasConceptScore W4285294350C15744967 @default.
- W4285294350 hasConceptScore W4285294350C162324750 @default.
- W4285294350 hasConceptScore W4285294350C173801870 @default.
- W4285294350 hasConceptScore W4285294350C175444787 @default.
- W4285294350 hasConceptScore W4285294350C180747234 @default.
- W4285294350 hasConceptScore W4285294350C19417346 @default.
- W4285294350 hasConceptScore W4285294350C2778869765 @default.
- W4285294350 hasConceptScore W4285294350C41008148 @default.
- W4285294350 hasConceptScore W4285294350C47177190 @default.
- W4285294350 hasConceptScore W4285294350C80444323 @default.
- W4285294350 hasConceptScore W4285294350C97541855 @default.
- W4285294350 hasFunder F4320338336 @default.
- W4285294350 hasIssue "3" @default.
- W4285294350 hasLocation W42852943501 @default.
- W4285294350 hasOpenAccess W4285294350 @default.
- W4285294350 hasPrimaryLocation W42852943501 @default.
- W4285294350 hasRelatedWork W2587229774 @default.
- W4285294350 hasRelatedWork W2890406131 @default.
- W4285294350 hasRelatedWork W2899084033 @default.
- W4285294350 hasRelatedWork W3012552522 @default.
- W4285294350 hasRelatedWork W3022038857 @default.
- W4285294350 hasRelatedWork W3173051288 @default.
- W4285294350 hasRelatedWork W3197854638 @default.
- W4285294350 hasRelatedWork W4226336685 @default.
- W4285294350 hasRelatedWork W4295352814 @default.
- W4285294350 hasRelatedWork W4319083788 @default.
- W4285294350 hasVolume "70" @default.
- W4285294350 isParatext "false" @default.
- W4285294350 isRetracted "false" @default.
- W4285294350 workType "article" @default.