Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213127955> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W3213127955 abstract "In reinforcement learning, continuous time is often discretized by a time scale $delta$, to which the resulting performance is known to be highly sensitive. In this work, we seek to find a $delta$-invariant algorithm for policy gradient (PG) methods, which performs well regardless of the value of $delta$. We first identify the underlying reasons that cause PG methods to fail as $delta to 0$, proving that the variance of the PG estimator can diverge to infinity in stochastic environments under a certain assumption of stochasticity. While durative actions or action repetition can be employed to have $delta$-invariance, previous action repetition methods cannot immediately react to unexpected situations in stochastic environments. We thus propose a novel $delta$-invariant method named Safe Action Repetition (SAR) applicable to any existing PG algorithm. SAR can handle the stochasticity of environments by adaptively reacting to changes in states during action repetition. We empirically show that our method is not only $delta$-invariant but also robust to stochasticity, outperforming previous $delta$-invariant approaches on eight MuJoCo environments with both deterministic and stochastic settings. Our code is available at https://vision.snu.ac.kr/projects/sar." @default.
- W3213127955 created "2021-11-22" @default.
- W3213127955 creator A5006868457 @default.
- W3213127955 creator A5032124306 @default.
- W3213127955 creator A5049929168 @default.
- W3213127955 date "2021-11-06" @default.
- W3213127955 modified "2023-09-27" @default.
- W3213127955 title "Time Discretization-Invariant Safe Action Repetition for Policy Gradient Methods" @default.
- W3213127955 doi "https://doi.org/10.48550/arxiv.2111.03941" @default.
- W3213127955 hasPublicationYear "2021" @default.
- W3213127955 type Work @default.
- W3213127955 sameAs 3213127955 @default.
- W3213127955 citedByCount "0" @default.
- W3213127955 crossrefType "posted-content" @default.
- W3213127955 hasAuthorship W3213127955A5006868457 @default.
- W3213127955 hasAuthorship W3213127955A5032124306 @default.
- W3213127955 hasAuthorship W3213127955A5049929168 @default.
- W3213127955 hasBestOaLocation W32131279551 @default.
- W3213127955 hasConcept C105795698 @default.
- W3213127955 hasConcept C11413529 @default.
- W3213127955 hasConcept C121332964 @default.
- W3213127955 hasConcept C126255220 @default.
- W3213127955 hasConcept C134306372 @default.
- W3213127955 hasConcept C154945302 @default.
- W3213127955 hasConcept C185429906 @default.
- W3213127955 hasConcept C190470478 @default.
- W3213127955 hasConcept C2780791683 @default.
- W3213127955 hasConcept C28826006 @default.
- W3213127955 hasConcept C33923547 @default.
- W3213127955 hasConcept C37914503 @default.
- W3213127955 hasConcept C41008148 @default.
- W3213127955 hasConcept C62520636 @default.
- W3213127955 hasConcept C73000952 @default.
- W3213127955 hasConcept C97541855 @default.
- W3213127955 hasConceptScore W3213127955C105795698 @default.
- W3213127955 hasConceptScore W3213127955C11413529 @default.
- W3213127955 hasConceptScore W3213127955C121332964 @default.
- W3213127955 hasConceptScore W3213127955C126255220 @default.
- W3213127955 hasConceptScore W3213127955C134306372 @default.
- W3213127955 hasConceptScore W3213127955C154945302 @default.
- W3213127955 hasConceptScore W3213127955C185429906 @default.
- W3213127955 hasConceptScore W3213127955C190470478 @default.
- W3213127955 hasConceptScore W3213127955C2780791683 @default.
- W3213127955 hasConceptScore W3213127955C28826006 @default.
- W3213127955 hasConceptScore W3213127955C33923547 @default.
- W3213127955 hasConceptScore W3213127955C37914503 @default.
- W3213127955 hasConceptScore W3213127955C41008148 @default.
- W3213127955 hasConceptScore W3213127955C62520636 @default.
- W3213127955 hasConceptScore W3213127955C73000952 @default.
- W3213127955 hasConceptScore W3213127955C97541855 @default.
- W3213127955 hasLocation W32131279551 @default.
- W3213127955 hasOpenAccess W3213127955 @default.
- W3213127955 hasPrimaryLocation W32131279551 @default.
- W3213127955 hasRelatedWork W1978704752 @default.
- W3213127955 hasRelatedWork W1986679470 @default.
- W3213127955 hasRelatedWork W1999189895 @default.
- W3213127955 hasRelatedWork W2032660751 @default.
- W3213127955 hasRelatedWork W2152704622 @default.
- W3213127955 hasRelatedWork W2323140291 @default.
- W3213127955 hasRelatedWork W2805040974 @default.
- W3213127955 hasRelatedWork W2963110188 @default.
- W3213127955 hasRelatedWork W4303104914 @default.
- W3213127955 hasRelatedWork W2556812935 @default.
- W3213127955 isParatext "false" @default.
- W3213127955 isRetracted "false" @default.
- W3213127955 magId "3213127955" @default.
- W3213127955 workType "article" @default.