Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890597452> ?p ?o ?g. }
- W2890597452 endingPage "5454" @default.
- W2890597452 startingPage "5442" @default.
- W2890597452 abstract "Policy optimization is an effective reinforcement learning approach to solve continuous control tasks. Recent achievements have shown that alternating online and offline optimization is a successful choice for efficient trajectory reuse. However, deciding when to stop optimizing and collect new trajectories is non-trivial, as it requires to account for the variance of the objective function estimate. In this paper, we propose a novel, model-free, policy search algorithm, POIS, applicable in both action-based and parameter-based settings. We first derive a high-confidence bound for importance sampling estimation; then we define a surrogate objective function, which is optimized offline whenever a new batch of trajectories is collected. Finally, the algorithm is tested on a selection of continuous control tasks, with both linear and deep policies, and compared with state-of-the-art policy optimization methods." @default.
- W2890597452 created "2018-09-27" @default.
- W2890597452 creator A5013289052 @default.
- W2890597452 creator A5015303026 @default.
- W2890597452 creator A5017130830 @default.
- W2890597452 creator A5037963323 @default.
- W2890597452 date "2018-01-01" @default.
- W2890597452 modified "2023-09-27" @default.
- W2890597452 title "Policy optimization via importance sampling" @default.
- W2890597452 hasPublicationYear "2018" @default.
- W2890597452 type Work @default.
- W2890597452 sameAs 2890597452 @default.
- W2890597452 citedByCount "24" @default.
- W2890597452 countsByYear W28905974522019 @default.
- W2890597452 countsByYear W28905974522020 @default.
- W2890597452 countsByYear W28905974522021 @default.
- W2890597452 countsByYear W28905974522022 @default.
- W2890597452 crossrefType "proceedings-article" @default.
- W2890597452 hasAuthorship W2890597452A5013289052 @default.
- W2890597452 hasAuthorship W2890597452A5015303026 @default.
- W2890597452 hasAuthorship W2890597452A5017130830 @default.
- W2890597452 hasAuthorship W2890597452A5037963323 @default.
- W2890597452 hasConcept C106131492 @default.
- W2890597452 hasConcept C11413529 @default.
- W2890597452 hasConcept C119857082 @default.
- W2890597452 hasConcept C121332964 @default.
- W2890597452 hasConcept C121955636 @default.
- W2890597452 hasConcept C126255220 @default.
- W2890597452 hasConcept C127413603 @default.
- W2890597452 hasConcept C1276947 @default.
- W2890597452 hasConcept C13662910 @default.
- W2890597452 hasConcept C137836250 @default.
- W2890597452 hasConcept C14036430 @default.
- W2890597452 hasConcept C140779682 @default.
- W2890597452 hasConcept C144133560 @default.
- W2890597452 hasConcept C154945302 @default.
- W2890597452 hasConcept C173246807 @default.
- W2890597452 hasConcept C196083921 @default.
- W2890597452 hasConcept C196921405 @default.
- W2890597452 hasConcept C206588197 @default.
- W2890597452 hasConcept C2775924081 @default.
- W2890597452 hasConcept C31972630 @default.
- W2890597452 hasConcept C33923547 @default.
- W2890597452 hasConcept C41008148 @default.
- W2890597452 hasConcept C48103436 @default.
- W2890597452 hasConcept C548081761 @default.
- W2890597452 hasConcept C78458016 @default.
- W2890597452 hasConcept C81917197 @default.
- W2890597452 hasConcept C86803240 @default.
- W2890597452 hasConcept C91575142 @default.
- W2890597452 hasConcept C97541855 @default.
- W2890597452 hasConceptScore W2890597452C106131492 @default.
- W2890597452 hasConceptScore W2890597452C11413529 @default.
- W2890597452 hasConceptScore W2890597452C119857082 @default.
- W2890597452 hasConceptScore W2890597452C121332964 @default.
- W2890597452 hasConceptScore W2890597452C121955636 @default.
- W2890597452 hasConceptScore W2890597452C126255220 @default.
- W2890597452 hasConceptScore W2890597452C127413603 @default.
- W2890597452 hasConceptScore W2890597452C1276947 @default.
- W2890597452 hasConceptScore W2890597452C13662910 @default.
- W2890597452 hasConceptScore W2890597452C137836250 @default.
- W2890597452 hasConceptScore W2890597452C14036430 @default.
- W2890597452 hasConceptScore W2890597452C140779682 @default.
- W2890597452 hasConceptScore W2890597452C144133560 @default.
- W2890597452 hasConceptScore W2890597452C154945302 @default.
- W2890597452 hasConceptScore W2890597452C173246807 @default.
- W2890597452 hasConceptScore W2890597452C196083921 @default.
- W2890597452 hasConceptScore W2890597452C196921405 @default.
- W2890597452 hasConceptScore W2890597452C206588197 @default.
- W2890597452 hasConceptScore W2890597452C2775924081 @default.
- W2890597452 hasConceptScore W2890597452C31972630 @default.
- W2890597452 hasConceptScore W2890597452C33923547 @default.
- W2890597452 hasConceptScore W2890597452C41008148 @default.
- W2890597452 hasConceptScore W2890597452C48103436 @default.
- W2890597452 hasConceptScore W2890597452C548081761 @default.
- W2890597452 hasConceptScore W2890597452C78458016 @default.
- W2890597452 hasConceptScore W2890597452C81917197 @default.
- W2890597452 hasConceptScore W2890597452C86803240 @default.
- W2890597452 hasConceptScore W2890597452C91575142 @default.
- W2890597452 hasConceptScore W2890597452C97541855 @default.
- W2890597452 hasLocation W28905974521 @default.
- W2890597452 hasOpenAccess W2890597452 @default.
- W2890597452 hasPrimaryLocation W28905974521 @default.
- W2890597452 hasRelatedWork W1547925194 @default.
- W2890597452 hasRelatedWork W1771410628 @default.
- W2890597452 hasRelatedWork W2012587148 @default.
- W2890597452 hasRelatedWork W2107438106 @default.
- W2890597452 hasRelatedWork W2111355007 @default.
- W2890597452 hasRelatedWork W2119567691 @default.
- W2890597452 hasRelatedWork W2119717200 @default.
- W2890597452 hasRelatedWork W2121863487 @default.
- W2890597452 hasRelatedWork W2125612430 @default.
- W2890597452 hasRelatedWork W2145339207 @default.
- W2890597452 hasRelatedWork W2155027007 @default.
- W2890597452 hasRelatedWork W2165150801 @default.
- W2890597452 hasRelatedWork W2594203335 @default.
- W2890597452 hasRelatedWork W2767133776 @default.
- W2890597452 hasRelatedWork W2890022552 @default.