FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2916826721> ?p ?o ?g. }

• W2916826721 abstract "Deep Reinforcement Learning has shown great success in a variety of control tasks. However, it is unclear how close we are to the vision of putting Deep RL into practice to solve real world problems. In particular, common practice in the field is to train policies on largely deterministic simulators and to evaluate algorithms through training performance alone, without a train/test distinction to ensure models generalise and are not overfitted. Moreover, it is not standard practice to check for generalisation under domain shift, although robustness to such system change between training and testing would be necessary for real-world Deep RL control, for example, in robotics. In this paper we study these issues by first characterising the sources of uncertainty that provide generalisation challenges in Deep RL. We then provide a new benchmark and thorough empirical evaluation of generalisation challenges for state of the art Deep RL methods. In particular, we show that, if generalisation is the goal, then common practice of evaluating algorithms based on their training performance leads to the wrong conclusions about algorithm choice. Finally, we evaluate several techniques for improving generalisation and draw conclusions about the most robust techniques to date." @default.
• W2916826721 created "2019-03-02" @default.
• W2916826721 creator A5017689065 @default.
• W2916826721 creator A5039838520 @default.
• W2916826721 creator A5042850624 @default.
• W2916826721 creator A5087823932 @default.
• W2916826721 date "2019-02-19" @default.
• W2916826721 modified "2023-09-27" @default.
• W2916826721 title "Investigating Generalisation in Continuous Deep Reinforcement Learning" @default.
• W2916826721 cites W1978161072 @default.
• W2916826721 cites W2145339207 @default.
• W2916826721 cites W2158782408 @default.
• W2916826721 cites W2169136412 @default.
• W2916826721 cites W2342662072 @default.
• W2916826721 cites W2602963933 @default.
• W2916826721 cites W2736601468 @default.
• W2916826721 cites W2756202949 @default.
• W2916826721 cites W2766447205 @default.
• W2916826721 cites W2773691349 @default.
• W2916826721 cites W2781585732 @default.
• W2916826721 cites W2797527950 @default.
• W2916826721 cites W2809256243 @default.
• W2916826721 cites W2809668646 @default.
• W2916826721 cites W2898436992 @default.
• W2916826721 cites W2903181768 @default.
• W2916826721 cites W2949103145 @default.
• W2916826721 cites W2949608212 @default.
• W2916826721 cites W2963120839 @default.
• W2916826721 cites W2963864421 @default.
• W2916826721 cites W2963973314 @default.
• W2916826721 cites W2964108292 @default.
• W2916826721 cites W2964248288 @default.
• W2916826721 cites W3093329015 @default.
• W2916826721 cites W64088143 @default.
• W2916826721 hasPublicationYear "2019" @default.
• W2916826721 type Work @default.
• W2916826721 sameAs 2916826721 @default.
• W2916826721 citedByCount "21" @default.
• W2916826721 countsByYear W29168267212019 @default.
• W2916826721 countsByYear W29168267212020 @default.
• W2916826721 countsByYear W29168267212021 @default.
• W2916826721 crossrefType "posted-content" @default.
• W2916826721 hasAuthorship W2916826721A5017689065 @default.
• W2916826721 hasAuthorship W2916826721A5039838520 @default.
• W2916826721 hasAuthorship W2916826721A5042850624 @default.
• W2916826721 hasAuthorship W2916826721A5087823932 @default.
• W2916826721 hasConcept C104317684 @default.
• W2916826721 hasConcept C108583219 @default.
• W2916826721 hasConcept C119857082 @default.
• W2916826721 hasConcept C13280743 @default.
• W2916826721 hasConcept C136197465 @default.
• W2916826721 hasConcept C154945302 @default.
• W2916826721 hasConcept C185592680 @default.
• W2916826721 hasConcept C185798385 @default.
• W2916826721 hasConcept C202444582 @default.
• W2916826721 hasConcept C205649164 @default.
• W2916826721 hasConcept C2775924081 @default.
• W2916826721 hasConcept C33923547 @default.
• W2916826721 hasConcept C34413123 @default.
• W2916826721 hasConcept C41008148 @default.
• W2916826721 hasConcept C55493867 @default.
• W2916826721 hasConcept C63479239 @default.
• W2916826721 hasConcept C90509273 @default.
• W2916826721 hasConcept C9652623 @default.
• W2916826721 hasConcept C97541855 @default.
• W2916826721 hasConceptScore W2916826721C104317684 @default.
• W2916826721 hasConceptScore W2916826721C108583219 @default.
• W2916826721 hasConceptScore W2916826721C119857082 @default.
• W2916826721 hasConceptScore W2916826721C13280743 @default.
• W2916826721 hasConceptScore W2916826721C136197465 @default.
• W2916826721 hasConceptScore W2916826721C154945302 @default.
• W2916826721 hasConceptScore W2916826721C185592680 @default.
• W2916826721 hasConceptScore W2916826721C185798385 @default.
• W2916826721 hasConceptScore W2916826721C202444582 @default.
• W2916826721 hasConceptScore W2916826721C205649164 @default.
• W2916826721 hasConceptScore W2916826721C2775924081 @default.
• W2916826721 hasConceptScore W2916826721C33923547 @default.
• W2916826721 hasConceptScore W2916826721C34413123 @default.
• W2916826721 hasConceptScore W2916826721C41008148 @default.
• W2916826721 hasConceptScore W2916826721C55493867 @default.
• W2916826721 hasConceptScore W2916826721C63479239 @default.
• W2916826721 hasConceptScore W2916826721C90509273 @default.
• W2916826721 hasConceptScore W2916826721C9652623 @default.
• W2916826721 hasConceptScore W2916826721C97541855 @default.
• W2916826721 hasLocation W29168267211 @default.
• W2916826721 hasOpenAccess W2916826721 @default.
• W2916826721 hasPrimaryLocation W29168267211 @default.
• W2916826721 hasRelatedWork W1757796397 @default.
• W2916826721 hasRelatedWork W1771410628 @default.
• W2916826721 hasRelatedWork W2100677568 @default.
• W2916826721 hasRelatedWork W2104228245 @default.
• W2916826721 hasRelatedWork W2145339207 @default.
• W2916826721 hasRelatedWork W2158782408 @default.
• W2916826721 hasRelatedWork W2257979135 @default.
• W2916826721 hasRelatedWork W2550182557 @default.
• W2916826721 hasRelatedWork W2602963933 @default.
• W2916826721 hasRelatedWork W2605102758 @default.
• W2916826721 hasRelatedWork W2736601468 @default.
• W2916826721 hasRelatedWork W2797527950 @default.
• W2916826721 hasRelatedWork W2809668646 @default.

• next