Matches in SemOpenAlex for { <https://semopenalex.org/work/W4200631568> ?p ?o ?g. }
- W4200631568 endingPage "45" @default.
- W4200631568 startingPage "45" @default.
- W4200631568 abstract "We present a deep machine learning (ML) approach to constraining cosmological parameters with multi-wavelength observations of galaxy clusters. The ML approach has two components: an encoder that builds a compressed representation of each galaxy cluster and a flexible CNN to estimate the cosmological model from a cluster sample. It is trained and tested on simulated cluster catalogs built from the Magneticum simulations. From the simulated catalogs, the ML method estimates the amplitude of matter fluctuations, sigma_8, at approximately the expected theoretical limit. More importantly, the deep ML approach can be interpreted. We lay out three schemes for interpreting the ML technique: a leave-one-out method for assessing cluster importance, an average saliency for evaluating feature importance, and correlations in the terse layer for understanding whether an ML technique can be safely applied to observational data. These interpretation schemes led to the discovery of a previously unknown self-calibration mode for flux- and volume-limited cluster surveys. We describe this new mode, which uses the amplitude and peak of the cluster mass PDF as anchors for mass calibration. We introduce the term overspecialized to describe a common pitfall in astronomical applications of machine learning in which the ML method learns simulation-specific details, and we show how a carefully constructed architecture can be used to check for this source of systematic error." @default.
- W4200631568 created "2021-12-31" @default.
- W4200631568 creator A5009547264 @default.
- W4200631568 creator A5070787181 @default.
- W4200631568 date "2022-02-01" @default.
- W4200631568 modified "2023-09-28" @default.
- W4200631568 title "The Importance of Being Interpretable: Toward an Understandable Machine Learning Encoder for Galaxy Cluster Cosmology" @default.
- W4200631568 cites W1552290780 @default.
- W4200631568 cites W1603209501 @default.
- W4200631568 cites W1846958989 @default.
- W4200631568 cites W1850774354 @default.
- W4200631568 cites W1919817864 @default.
- W4200631568 cites W1978193295 @default.
- W4200631568 cites W1979127253 @default.
- W4200631568 cites W1979535378 @default.
- W4200631568 cites W1990665624 @default.
- W4200631568 cites W1990930372 @default.
- W4200631568 cites W1997254864 @default.
- W4200631568 cites W1997283336 @default.
- W4200631568 cites W1998515260 @default.
- W4200631568 cites W2007772592 @default.
- W4200631568 cites W2028704213 @default.
- W4200631568 cites W2030605651 @default.
- W4200631568 cites W2034255160 @default.
- W4200631568 cites W2036370896 @default.
- W4200631568 cites W2044738244 @default.
- W4200631568 cites W2050699133 @default.
- W4200631568 cites W2055130152 @default.
- W4200631568 cites W2059330744 @default.
- W4200631568 cites W2077353758 @default.
- W4200631568 cites W2085061516 @default.
- W4200631568 cites W2096825304 @default.
- W4200631568 cites W2098564368 @default.
- W4200631568 cites W2100514692 @default.
- W4200631568 cites W2100837042 @default.
- W4200631568 cites W2104900201 @default.
- W4200631568 cites W2141968105 @default.
- W4200631568 cites W2147875900 @default.
- W4200631568 cites W2170857649 @default.
- W4200631568 cites W2177551589 @default.
- W4200631568 cites W2246040896 @default.
- W4200631568 cites W2253699394 @default.
- W4200631568 cites W2265168320 @default.
- W4200631568 cites W2301117188 @default.
- W4200631568 cites W2538960942 @default.
- W4200631568 cites W2616298216 @default.
- W4200631568 cites W2751739721 @default.
- W4200631568 cites W2768653164 @default.
- W4200631568 cites W2823209005 @default.
- W4200631568 cites W2945976633 @default.
- W4200631568 cites W2964716939 @default.
- W4200631568 cites W2969096242 @default.
- W4200631568 cites W2998594699 @default.
- W4200631568 cites W3023429516 @default.
- W4200631568 cites W3041488807 @default.
- W4200631568 cites W3098169525 @default.
- W4200631568 cites W3098793509 @default.
- W4200631568 cites W3099750564 @default.
- W4200631568 cites W3100294950 @default.
- W4200631568 cites W3100414533 @default.
- W4200631568 cites W3100574878 @default.
- W4200631568 cites W3100711616 @default.
- W4200631568 cites W3101424783 @default.
- W4200631568 cites W3102844577 @default.
- W4200631568 cites W3102887027 @default.
- W4200631568 cites W3103290085 @default.
- W4200631568 cites W3104786947 @default.
- W4200631568 cites W3105595725 @default.
- W4200631568 cites W3106091538 @default.
- W4200631568 cites W3106152809 @default.
- W4200631568 cites W3115503526 @default.
- W4200631568 cites W3122008711 @default.
- W4200631568 cites W3129156620 @default.
- W4200631568 doi "https://doi.org/10.3847/1538-4357/ac423e" @default.
- W4200631568 hasPublicationYear "2022" @default.
- W4200631568 type Work @default.
- W4200631568 citedByCount "3" @default.
- W4200631568 countsByYear W42006315682022 @default.
- W4200631568 countsByYear W42006315682023 @default.
- W4200631568 crossrefType "journal-article" @default.
- W4200631568 hasAuthorship W4200631568A5009547264 @default.
- W4200631568 hasAuthorship W4200631568A5070787181 @default.
- W4200631568 hasBestOaLocation W42006315681 @default.
- W4200631568 hasConcept C119857082 @default.
- W4200631568 hasConcept C120665830 @default.
- W4200631568 hasConcept C121332964 @default.
- W4200631568 hasConcept C138885662 @default.
- W4200631568 hasConcept C153180895 @default.
- W4200631568 hasConcept C154945302 @default.
- W4200631568 hasConcept C156055797 @default.
- W4200631568 hasConcept C159249277 @default.
- W4200631568 hasConcept C164866538 @default.
- W4200631568 hasConcept C165838908 @default.
- W4200631568 hasConcept C180205008 @default.
- W4200631568 hasConcept C199360897 @default.
- W4200631568 hasConcept C26405456 @default.
- W4200631568 hasConcept C2776401178 @default.
- W4200631568 hasConcept C41008148 @default.