Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385852790> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4385852790 endingPage "3194" @default.
- W4385852790 startingPage "3181" @default.
- W4385852790 abstract "ABSTRACT We introduce a novel approach to boost the efficiency of the importance nested sampling (INS) technique for Bayesian posterior and evidence estimation using deep learning. Unlike rejection-based sampling methods such as vanilla nested sampling (NS) or Markov chain Monte Carlo (MCMC) algorithms, importance sampling techniques can use all likelihood evaluations for posterior and evidence estimation. However, for efficient importance sampling, one needs proposal distributions that closely mimic the posterior distributions. We show how to combine INS with deep learning via neural network regression to accomplish this task. We also introduce nautilus, a reference open-source python implementation of this technique for Bayesian posterior and evidence estimation. We compare nautilus against popular NS and MCMC packages, including emcee, dynesty, ultranest, and pocomc, on a variety of challenging synthetic problems and real-world applications in exoplanet detection, galaxy SED fitting and cosmology. In all applications, the sampling efficiency of nautilus is substantially higher than that of all other samplers, often by more than an order of magnitude. Simultaneously, nautilus delivers highly accurate results and needs fewer likelihood evaluations than all other samplers tested. We also show that nautilus has good scaling with the dimensionality of the likelihood and is easily parallelizable to many CPUs." @default.
- W4385852790 created "2023-08-17" @default.
- W4385852790 creator A5038103942 @default.
- W4385852790 date "2023-08-16" @default.
- W4385852790 modified "2023-09-27" @default.
- W4385852790 title "<scp>nautilus</scp>: boosting Bayesian importance nested sampling with deep learning" @default.
- W4385852790 cites W1607306679 @default.
- W4385852790 cites W1836665496 @default.
- W4385852790 cites W1886713477 @default.
- W4385852790 cites W1897231400 @default.
- W4385852790 cites W2011301426 @default.
- W4385852790 cites W2043235658 @default.
- W4385852790 cites W2135625048 @default.
- W4385852790 cites W2146292423 @default.
- W4385852790 cites W2172939019 @default.
- W4385852790 cites W2463842489 @default.
- W4385852790 cites W2607418782 @default.
- W4385852790 cites W2772961215 @default.
- W4385852790 cites W2921963248 @default.
- W4385852790 cites W2923635089 @default.
- W4385852790 cites W2931388211 @default.
- W4385852790 cites W2964121744 @default.
- W4385852790 cites W2972133452 @default.
- W4385852790 cites W2991395895 @default.
- W4385852790 cites W3099210727 @default.
- W4385852790 cites W3102014803 @default.
- W4385852790 cites W3104188978 @default.
- W4385852790 cites W3104579003 @default.
- W4385852790 cites W3104624969 @default.
- W4385852790 cites W3106171379 @default.
- W4385852790 cites W3132521674 @default.
- W4385852790 cites W3174865733 @default.
- W4385852790 cites W3194244178 @default.
- W4385852790 cites W32980360 @default.
- W4385852790 cites W4221153697 @default.
- W4385852790 cites W4221161143 @default.
- W4385852790 cites W4231491688 @default.
- W4385852790 cites W4281552525 @default.
- W4385852790 cites W4292155219 @default.
- W4385852790 cites W4293312513 @default.
- W4385852790 cites W4304014078 @default.
- W4385852790 cites W4308654784 @default.
- W4385852790 cites W4321441607 @default.
- W4385852790 doi "https://doi.org/10.1093/mnras/stad2441" @default.
- W4385852790 hasPublicationYear "2023" @default.
- W4385852790 type Work @default.
- W4385852790 citedByCount "0" @default.
- W4385852790 crossrefType "journal-article" @default.
- W4385852790 hasAuthorship W4385852790A5038103942 @default.
- W4385852790 hasBestOaLocation W43858527902 @default.
- W4385852790 hasConcept C106131492 @default.
- W4385852790 hasConcept C107673813 @default.
- W4385852790 hasConcept C111350023 @default.
- W4385852790 hasConcept C119857082 @default.
- W4385852790 hasConcept C140779682 @default.
- W4385852790 hasConcept C154945302 @default.
- W4385852790 hasConcept C158424031 @default.
- W4385852790 hasConcept C31972630 @default.
- W4385852790 hasConcept C41008148 @default.
- W4385852790 hasConcept C95923904 @default.
- W4385852790 hasConceptScore W4385852790C106131492 @default.
- W4385852790 hasConceptScore W4385852790C107673813 @default.
- W4385852790 hasConceptScore W4385852790C111350023 @default.
- W4385852790 hasConceptScore W4385852790C119857082 @default.
- W4385852790 hasConceptScore W4385852790C140779682 @default.
- W4385852790 hasConceptScore W4385852790C154945302 @default.
- W4385852790 hasConceptScore W4385852790C158424031 @default.
- W4385852790 hasConceptScore W4385852790C31972630 @default.
- W4385852790 hasConceptScore W4385852790C41008148 @default.
- W4385852790 hasConceptScore W4385852790C95923904 @default.
- W4385852790 hasFunder F4320306076 @default.
- W4385852790 hasFunder F4320306084 @default.
- W4385852790 hasFunder F4320306101 @default.
- W4385852790 hasIssue "2" @default.
- W4385852790 hasLocation W43858527901 @default.
- W4385852790 hasLocation W43858527902 @default.
- W4385852790 hasOpenAccess W4385852790 @default.
- W4385852790 hasPrimaryLocation W43858527901 @default.
- W4385852790 hasRelatedWork W2117957819 @default.
- W4385852790 hasRelatedWork W2146501959 @default.
- W4385852790 hasRelatedWork W2172275095 @default.
- W4385852790 hasRelatedWork W2255115219 @default.
- W4385852790 hasRelatedWork W2352563150 @default.
- W4385852790 hasRelatedWork W2753217981 @default.
- W4385852790 hasRelatedWork W2940690269 @default.
- W4385852790 hasRelatedWork W3121470121 @default.
- W4385852790 hasRelatedWork W3198356641 @default.
- W4385852790 hasRelatedWork W2740715953 @default.
- W4385852790 hasVolume "525" @default.
- W4385852790 isParatext "false" @default.
- W4385852790 isRetracted "false" @default.
- W4385852790 workType "article" @default.