Understanding Deep Learning with Statistical Relevance.
This paper argues that a notion of statistical explanation, based on Salmon's statistical relevance model, can help us better understand deep neural networks. It is proved that homogeneous partitions, the core notion of Salmon's model, are equivalent to minimal sufficient statistics, an important no...
| Publicado en: | Philosophy of Science Vol. 89; no. 1; pp. 20 - 42 |
|---|---|
| Autor principal: | |
| Formato: | Artículo |
| Publicado: |
Cambridge University Press
Jan2022
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=159632386&site=ehost-live header: @attributes: shortDbName: hlh uiTerm: 159632386 longDbName: Humanities International Complete uiTag: AN controlInfo: bkinfo: jinfo: jid: 00318248 PSC jtl: Philosophy of Science issn: 00318248 maglogo: N pubinfo: dt: Jan2022 vid: 89 iid: 1 pid: 15979 pub: Cambridge University Press artinfo: ui: 159632386 10.1017/psa.2021.12 ppf: 20 ppct: 22 formats: fmt: – @attributes: type: T – @attributes: type: P size: 399KB tig: atl: Understanding Deep Learning with Statistical Relevance. aug: au: Räz, Tim affil: University of Bern, Institute of Philosophy, Bern, Switzerland su: Artificial neural networks Statistical learning Deep learning Inferential statistics Statistical models sug: subj: Artificial neural networks Statistical learning Deep learning Inferential statistics Statistical models ab: This paper argues that a notion of statistical explanation, based on Salmon's statistical relevance model, can help us better understand deep neural networks. It is proved that homogeneous partitions, the core notion of Salmon's model, are equivalent to minimal sufficient statistics, an important notion from statistical inference. This establishes a link to deep neural networks via the so-called Information Bottleneck method, an information-theoretic framework, according to which deep neural networks implicitly solve an optimization problem that generalizes minimal sufficient statistics. The resulting notion of statistical explanation is general, mathematical, and subcausal. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y custom: Copyright of Philosophy of Science is the property of Cambridge University Press and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. item: Philosophy of Science holder: Cambridge University Press dt: @attributes: year: 2022 holdings: @attributes: islocal: N |
|---|