Matches in SemOpenAlex for { <https://semopenalex.org/work/W2481567506> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W2481567506 abstract "This paper introduces a novel method for learning how to play the most difficult Atari 2600 games from the Arcade Learning Environment using deep reinforcement learning. The proposed method, human checkpoint replay, consists in using checkpoints sampled from human gameplay as starting points for the learning process. This is meant to compensate for the difficulties of current exploration strategies, such as epsilon-greedy, to find successful control policies in games with sparse rewards. Like other deep reinforcement learning architectures, our model uses a convolutional neural network that receives only raw pixel inputs to estimate the state value function. We tested our method on Montezuma's Revenge and Private Eye, two of the most challenging games from the Atari platform. The results we obtained show a substantial improvement compared to previous learning approaches, as well as over a random player. We also propose a method for training deep reinforcement learning agents using human gameplay experience, which we call human experience replay." @default.
- W2481567506 created "2016-08-23" @default.
- W2481567506 creator A5042558253 @default.
- W2481567506 creator A5077365712 @default.
- W2481567506 date "2016-07-18" @default.
- W2481567506 modified "2023-09-27" @default.
- W2481567506 title "Playing Atari Games with Deep Reinforcement Learning and Human Checkpoint Replay" @default.
- W2481567506 cites W1512866498 @default.
- W2481567506 cites W1595483645 @default.
- W2481567506 cites W1602154927 @default.
- W2481567506 cites W1658008008 @default.
- W2481567506 cites W1757796397 @default.
- W2481567506 cites W2099397840 @default.
- W2481567506 cites W2145339207 @default.
- W2481567506 cites W2159635464 @default.
- W2481567506 cites W2168992242 @default.
- W2481567506 cites W2194775991 @default.
- W2481567506 cites W2201581102 @default.
- W2481567506 cites W2257979135 @default.
- W2481567506 cites W2280163991 @default.
- W2481567506 cites W2312609093 @default.
- W2481567506 cites W2949640717 @default.
- W2481567506 cites W2952523895 @default.
- W2481567506 cites W3089091950 @default.
- W2481567506 hasPublicationYear "2016" @default.
- W2481567506 type Work @default.
- W2481567506 sameAs 2481567506 @default.
- W2481567506 citedByCount "33" @default.
- W2481567506 countsByYear W24815675062017 @default.
- W2481567506 countsByYear W24815675062018 @default.
- W2481567506 countsByYear W24815675062019 @default.
- W2481567506 countsByYear W24815675062020 @default.
- W2481567506 countsByYear W24815675062021 @default.
- W2481567506 crossrefType "posted-content" @default.
- W2481567506 hasAuthorship W2481567506A5042558253 @default.
- W2481567506 hasAuthorship W2481567506A5077365712 @default.
- W2481567506 hasConcept C108583219 @default.
- W2481567506 hasConcept C111919701 @default.
- W2481567506 hasConcept C14036430 @default.
- W2481567506 hasConcept C144237770 @default.
- W2481567506 hasConcept C14646407 @default.
- W2481567506 hasConcept C154945302 @default.
- W2481567506 hasConcept C188116033 @default.
- W2481567506 hasConcept C33923547 @default.
- W2481567506 hasConcept C41008148 @default.
- W2481567506 hasConcept C78458016 @default.
- W2481567506 hasConcept C81363708 @default.
- W2481567506 hasConcept C86803240 @default.
- W2481567506 hasConcept C97541855 @default.
- W2481567506 hasConcept C98045186 @default.
- W2481567506 hasConceptScore W2481567506C108583219 @default.
- W2481567506 hasConceptScore W2481567506C111919701 @default.
- W2481567506 hasConceptScore W2481567506C14036430 @default.
- W2481567506 hasConceptScore W2481567506C144237770 @default.
- W2481567506 hasConceptScore W2481567506C14646407 @default.
- W2481567506 hasConceptScore W2481567506C154945302 @default.
- W2481567506 hasConceptScore W2481567506C188116033 @default.
- W2481567506 hasConceptScore W2481567506C33923547 @default.
- W2481567506 hasConceptScore W2481567506C41008148 @default.
- W2481567506 hasConceptScore W2481567506C78458016 @default.
- W2481567506 hasConceptScore W2481567506C81363708 @default.
- W2481567506 hasConceptScore W2481567506C86803240 @default.
- W2481567506 hasConceptScore W2481567506C97541855 @default.
- W2481567506 hasConceptScore W2481567506C98045186 @default.
- W2481567506 hasLocation W24815675061 @default.
- W2481567506 hasOpenAccess W2481567506 @default.
- W2481567506 hasPrimaryLocation W24815675061 @default.
- W2481567506 hasRelatedWork W1757796397 @default.
- W2481567506 hasRelatedWork W1999874108 @default.
- W2481567506 hasRelatedWork W2121863487 @default.
- W2481567506 hasRelatedWork W2145339207 @default.
- W2481567506 hasRelatedWork W2148112459 @default.
- W2481567506 hasRelatedWork W2155968351 @default.
- W2481567506 hasRelatedWork W2167224731 @default.
- W2481567506 hasRelatedWork W2173564293 @default.
- W2481567506 hasRelatedWork W2257979135 @default.
- W2481567506 hasRelatedWork W2415726935 @default.
- W2481567506 hasRelatedWork W2434014514 @default.
- W2481567506 hasRelatedWork W2736601468 @default.
- W2481567506 hasRelatedWork W2741122588 @default.
- W2481567506 hasRelatedWork W2962957031 @default.
- W2481567506 hasRelatedWork W2963094133 @default.
- W2481567506 hasRelatedWork W2963099939 @default.
- W2481567506 hasRelatedWork W2963376229 @default.
- W2481567506 hasRelatedWork W2963477884 @default.
- W2481567506 hasRelatedWork W2964043796 @default.
- W2481567506 hasRelatedWork W3103780890 @default.
- W2481567506 isParatext "false" @default.
- W2481567506 isRetracted "false" @default.
- W2481567506 magId "2481567506" @default.
- W2481567506 workType "article" @default.