Matches in SemOpenAlex for { <https://semopenalex.org/work/W3211730526> ?p ?o ?g. }
- W3211730526 abstract "Parametric stochastic simulators are ubiquitous in science, often featuring high-dimensional input parameters and/or an intractable likelihood. Performing Bayesian parameter inference in this context can be challenging. We present a neural simulator-based inference algorithm which simultaneously offers simulation efficiency and fast empirical posterior testability, which is unique among modern algorithms. Our approach is simulation efficient by simultaneously estimating low-dimensional marginal posteriors instead of the joint posterior and by proposing simulations targeted to an observation of interest via a prior suitably truncated by an indicator function. Furthermore, by estimating a locally amortized posterior our algorithm enables efficient empirical tests of the robustness of the inference results. Such tests are important for sanity-checking inference in real-world applications, which do not feature a known ground truth. We perform experiments on a marginalized version of the simulation-based inference benchmark and two complex and narrow posteriors, highlighting the simulator efficiency of our algorithm as well as the quality of the estimated marginal posteriors. Implementation on GitHub." @default.
- W3211730526 created "2021-11-22" @default.
- W3211730526 creator A5005753034 @default.
- W3211730526 creator A5009608259 @default.
- W3211730526 creator A5022536484 @default.
- W3211730526 creator A5063214673 @default.
- W3211730526 creator A5070368389 @default.
- W3211730526 date "2021-12-06" @default.
- W3211730526 modified "2023-09-24" @default.
- W3211730526 title "Truncated Marginal Neural Ratio Estimation" @default.
- W3211730526 cites W114923250 @default.
- W3211730526 cites W1503398984 @default.
- W3211730526 cites W1506806321 @default.
- W3211730526 cites W1524012148 @default.
- W3211730526 cites W1555683961 @default.
- W3211730526 cites W1560724230 @default.
- W3211730526 cites W1663973292 @default.
- W3211730526 cites W1793259860 @default.
- W3211730526 cites W1836665496 @default.
- W3211730526 cites W1983452151 @default.
- W3211730526 cites W1990119892 @default.
- W3211730526 cites W1991209709 @default.
- W3211730526 cites W2011301426 @default.
- W3211730526 cites W2029164135 @default.
- W3211730526 cites W2040011853 @default.
- W3211730526 cites W2042296034 @default.
- W3211730526 cites W2043979479 @default.
- W3211730526 cites W2056760934 @default.
- W3211730526 cites W2067392831 @default.
- W3211730526 cites W2128947363 @default.
- W3211730526 cites W2135889985 @default.
- W3211730526 cites W2136133754 @default.
- W3211730526 cites W2138309709 @default.
- W3211730526 cites W2152246075 @default.
- W3211730526 cites W2156984627 @default.
- W3211730526 cites W2158131535 @default.
- W3211730526 cites W2212660284 @default.
- W3211730526 cites W2262573983 @default.
- W3211730526 cites W2342249984 @default.
- W3211730526 cites W2516678343 @default.
- W3211730526 cites W2530741948 @default.
- W3211730526 cites W2607418782 @default.
- W3211730526 cites W2731338647 @default.
- W3211730526 cites W2752620087 @default.
- W3211730526 cites W2796687148 @default.
- W3211730526 cites W2798398359 @default.
- W3211730526 cites W2895152177 @default.
- W3211730526 cites W2903675476 @default.
- W3211730526 cites W2945598669 @default.
- W3211730526 cites W2959696098 @default.
- W3211730526 cites W2963047245 @default.
- W3211730526 cites W2963475251 @default.
- W3211730526 cites W2963818159 @default.
- W3211730526 cites W2963822196 @default.
- W3211730526 cites W2963884690 @default.
- W3211730526 cites W2964000438 @default.
- W3211730526 cites W2964129402 @default.
- W3211730526 cites W2964250089 @default.
- W3211730526 cites W2964340499 @default.
- W3211730526 cites W2970971581 @default.
- W3211730526 cites W3031514878 @default.
- W3211730526 cites W3034706502 @default.
- W3211730526 cites W3086988147 @default.
- W3211730526 cites W3096831136 @default.
- W3211730526 cites W3099878876 @default.
- W3211730526 cites W3104188978 @default.
- W3211730526 cites W3105656913 @default.
- W3211730526 cites W3106756783 @default.
- W3211730526 cites W3108276424 @default.
- W3211730526 cites W3150635270 @default.
- W3211730526 cites W3150807214 @default.
- W3211730526 cites W3157094080 @default.
- W3211730526 cites W3205185019 @default.
- W3211730526 cites W4226024614 @default.
- W3211730526 cites W632087424 @default.
- W3211730526 doi "https://doi.org/10.5281/zenodo.5043707" @default.
- W3211730526 hasPublicationYear "2021" @default.
- W3211730526 type Work @default.
- W3211730526 sameAs 3211730526 @default.
- W3211730526 citedByCount "0" @default.
- W3211730526 crossrefType "proceedings-article" @default.
- W3211730526 hasAuthorship W3211730526A5005753034 @default.
- W3211730526 hasAuthorship W3211730526A5009608259 @default.
- W3211730526 hasAuthorship W3211730526A5022536484 @default.
- W3211730526 hasAuthorship W3211730526A5063214673 @default.
- W3211730526 hasAuthorship W3211730526A5070368389 @default.
- W3211730526 hasConcept C104317684 @default.
- W3211730526 hasConcept C105795698 @default.
- W3211730526 hasConcept C107673813 @default.
- W3211730526 hasConcept C11413529 @default.
- W3211730526 hasConcept C117251300 @default.
- W3211730526 hasConcept C119857082 @default.
- W3211730526 hasConcept C13280743 @default.
- W3211730526 hasConcept C154945302 @default.
- W3211730526 hasConcept C160234255 @default.
- W3211730526 hasConcept C185592680 @default.
- W3211730526 hasConcept C185798385 @default.
- W3211730526 hasConcept C205649164 @default.
- W3211730526 hasConcept C2776214188 @default.
- W3211730526 hasConcept C33923547 @default.