Matches in SemOpenAlex for { <https://semopenalex.org/work/W3128466077> ?p ?o ?g. }
- W3128466077 abstract "We propose a novel learning paradigm, Self-Imitation via Reduction (SIR), for solving compositional reinforcement learning problems. SIR is based on two core ideas: task reduction and self-imitation. Task reduction tackles a hard-to-solve task by actively reducing it to an easier task whose solution is known by the RL agent. Once the original hard task is successfully solved by task reduction, the agent naturally obtains a self-generated solution trajectory to imitate. By continuously collecting and imitating such demonstrations, the agent is able to progressively expand the solved subspace in the entire task space. Experiment results show that SIR can significantly accelerate and improve learning on a variety of challenging sparse-reward continuous-control problems with compositional structures." @default.
- W3128466077 created "2021-02-15" @default.
- W3128466077 creator A5008951080 @default.
- W3128466077 creator A5009962102 @default.
- W3128466077 creator A5049093671 @default.
- W3128466077 creator A5066028215 @default.
- W3128466077 creator A5090856797 @default.
- W3128466077 date "2021-05-03" @default.
- W3128466077 modified "2023-10-16" @default.
- W3128466077 title "Solving Compositional Reinforcement Learning Problems via Task Reduction" @default.
- W3128466077 cites W1507087299 @default.
- W3128466077 cites W1538131130 @default.
- W3128466077 cites W1538393421 @default.
- W3128466077 cites W1569788011 @default.
- W3128466077 cites W1585861384 @default.
- W3128466077 cites W1592847719 @default.
- W3128466077 cites W1963873191 @default.
- W3128466077 cites W2031727428 @default.
- W3128466077 cites W2056584142 @default.
- W3128466077 cites W2108535023 @default.
- W3128466077 cites W2109910161 @default.
- W3128466077 cites W2130535800 @default.
- W3128466077 cites W2131241448 @default.
- W3128466077 cites W2132083787 @default.
- W3128466077 cites W2158548602 @default.
- W3128466077 cites W2158782408 @default.
- W3128466077 cites W2160371091 @default.
- W3128466077 cites W2549293531 @default.
- W3128466077 cites W2553882142 @default.
- W3128466077 cites W2557449848 @default.
- W3128466077 cites W2594829461 @default.
- W3128466077 cites W2736601468 @default.
- W3128466077 cites W2753738274 @default.
- W3128466077 cites W2775536965 @default.
- W3128466077 cites W2803281228 @default.
- W3128466077 cites W2804010078 @default.
- W3128466077 cites W2890010095 @default.
- W3128466077 cites W2893841966 @default.
- W3128466077 cites W2903327785 @default.
- W3128466077 cites W2909335861 @default.
- W3128466077 cites W2922007426 @default.
- W3128466077 cites W2954700257 @default.
- W3128466077 cites W2962719460 @default.
- W3128466077 cites W2962887844 @default.
- W3128466077 cites W2962902376 @default.
- W3128466077 cites W2963262099 @default.
- W3128466077 cites W2963277051 @default.
- W3128466077 cites W2963286043 @default.
- W3128466077 cites W2963293881 @default.
- W3128466077 cites W2963403593 @default.
- W3128466077 cites W2963403868 @default.
- W3128466077 cites W2963411833 @default.
- W3128466077 cites W2963444224 @default.
- W3128466077 cites W2964001908 @default.
- W3128466077 cites W2964021598 @default.
- W3128466077 cites W2964062135 @default.
- W3128466077 cites W2964118020 @default.
- W3128466077 cites W2964118262 @default.
- W3128466077 cites W2964198579 @default.
- W3128466077 cites W2964227312 @default.
- W3128466077 cites W2964327384 @default.
- W3128466077 cites W2964342357 @default.
- W3128466077 cites W2970263828 @default.
- W3128466077 cites W2970377754 @default.
- W3128466077 cites W2970534317 @default.
- W3128466077 cites W2970720334 @default.
- W3128466077 cites W2970786335 @default.
- W3128466077 cites W2970948392 @default.
- W3128466077 cites W2970990801 @default.
- W3128466077 cites W2972758308 @default.
- W3128466077 cites W2981234930 @default.
- W3128466077 cites W2994847258 @default.
- W3128466077 cites W2995520132 @default.
- W3128466077 cites W2995636097 @default.
- W3128466077 cites W2996037775 @default.
- W3128466077 cites W2996125406 @default.
- W3128466077 cites W2996354442 @default.
- W3128466077 cites W3030163527 @default.
- W3128466077 cites W3031840745 @default.
- W3128466077 cites W3032077725 @default.
- W3128466077 cites W3032377877 @default.
- W3128466077 cites W3034445277 @default.
- W3128466077 cites W3046190872 @default.
- W3128466077 cites W3089482831 @default.
- W3128466077 cites W3118210634 @default.
- W3128466077 cites W567721252 @default.
- W3128466077 hasPublicationYear "2021" @default.
- W3128466077 type Work @default.
- W3128466077 sameAs 3128466077 @default.
- W3128466077 citedByCount "3" @default.
- W3128466077 countsByYear W31284660772021 @default.
- W3128466077 crossrefType "proceedings-article" @default.
- W3128466077 hasAuthorship W3128466077A5008951080 @default.
- W3128466077 hasAuthorship W3128466077A5009962102 @default.
- W3128466077 hasAuthorship W3128466077A5049093671 @default.
- W3128466077 hasAuthorship W3128466077A5066028215 @default.
- W3128466077 hasAuthorship W3128466077A5090856797 @default.
- W3128466077 hasConcept C111335779 @default.
- W3128466077 hasConcept C126388530 @default.
- W3128466077 hasConcept C127413603 @default.