FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3169063507> ?p ?o ?g. }

• W3169063507 abstract "In this work we introduce a new approach for identifiable non-linear ICA models. Recently there has been a renaissance in identifiability results in deep generative models, not least for non-linear ICA. These prior works, however, have assumed access to a sufficiently-informative auxiliary set of observations, denoted $mathbf{u}$. We show here how identifiability can be obtained in the absence of this side-information, rendering possible fully-unsupervised identifiable non-linear ICA. While previous theoretical results have established the impossibility of identifiable non-linear ICA in the presence of infinitely-flexible universal function approximators, here we rely on the intrinsically-finite modelling capacity of any particular chosen parameterisation of a deep generative model. In particular, we focus on generative models which perform clustering in their latent space -- a model structure which matches previous identifiable models, but with the learnt clustering providing a synthetic form of auxiliary information. We evaluate our proposals using VAEs, on synthetic and image datasets, and find that the learned clusterings function effectively: deep generative models with latent clusterings are empirically identifiable, to the same degree as models which rely on side information." @default.
• W3169063507 created "2021-06-22" @default.
• W3169063507 creator A5019067441 @default.
• W3169063507 creator A5072420270 @default.
• W3169063507 date "2021-06-09" @default.
• W3169063507 modified "2023-09-27" @default.
• W3169063507 title "I Don't Need $mathbf{u}$: Identifiable Non-Linear ICA Without Side Information" @default.
• W3169063507 cites W1480376833 @default.
• W3169063507 cites W1533861849 @default.
• W3169063507 cites W1663973292 @default.
• W3169063507 cites W1909320841 @default.
• W3169063507 cites W1959608418 @default.
• W3169063507 cites W2012762214 @default.
• W3169063507 cites W2025341678 @default.
• W3169063507 cites W2037757210 @default.
• W3169063507 cites W2088658556 @default.
• W3169063507 cites W2097089247 @default.
• W3169063507 cites W2125027820 @default.
• W3169063507 cites W2127218421 @default.
• W3169063507 cites W2138204974 @default.
• W3169063507 cites W2157988812 @default.
• W3169063507 cites W2222512263 @default.
• W3169063507 cites W2337374958 @default.
• W3169063507 cites W2587284713 @default.
• W3169063507 cites W2730106296 @default.
• W3169063507 cites W2753738274 @default.
• W3169063507 cites W2763081248 @default.
• W3169063507 cites W2798991696 @default.
• W3169063507 cites W2842511635 @default.
• W3169063507 cites W2883725317 @default.
• W3169063507 cites W2905986396 @default.
• W3169063507 cites W2944828972 @default.
• W3169063507 cites W2951541198 @default.
• W3169063507 cites W2962690557 @default.
• W3169063507 cites W2962964508 @default.
• W3169063507 cites W2963104724 @default.
• W3169063507 cites W2963633581 @default.
• W3169063507 cites W2963762683 @default.
• W3169063507 cites W2964127395 @default.
• W3169063507 cites W2964193753 @default.
• W3169063507 cites W2965354194 @default.
• W3169063507 cites W2968131493 @default.
• W3169063507 cites W2976351901 @default.
• W3169063507 cites W2977046142 @default.
• W3169063507 cites W2979473749 @default.
• W3169063507 cites W2992035660 @default.
• W3169063507 cites W2995181141 @default.
• W3169063507 cites W2999977565 @default.
• W3169063507 cites W3005680577 @default.
• W3169063507 cites W3007285736 @default.
• W3169063507 cites W3035524453 @default.
• W3169063507 cites W3037805286 @default.
• W3169063507 cites W3042571928 @default.
• W3169063507 cites W3095121901 @default.
• W3169063507 cites W3106428938 @default.
• W3169063507 cites W3159681767 @default.
• W3169063507 cites W3163602117 @default.
• W3169063507 cites W3166725635 @default.
• W3169063507 cites W3214039422 @default.
• W3169063507 hasPublicationYear "2021" @default.
• W3169063507 type Work @default.
• W3169063507 sameAs 3169063507 @default.
• W3169063507 citedByCount "0" @default.
• W3169063507 crossrefType "posted-content" @default.
• W3169063507 hasAuthorship W3169063507A5019067441 @default.
• W3169063507 hasAuthorship W3169063507A5072420270 @default.
• W3169063507 hasConcept C119857082 @default.
• W3169063507 hasConcept C122770356 @default.
• W3169063507 hasConcept C14036430 @default.
• W3169063507 hasConcept C153180895 @default.
• W3169063507 hasConcept C154945302 @default.
• W3169063507 hasConcept C167966045 @default.
• W3169063507 hasConcept C205711294 @default.
• W3169063507 hasConcept C39890363 @default.
• W3169063507 hasConcept C41008148 @default.
• W3169063507 hasConcept C51167844 @default.
• W3169063507 hasConcept C73555534 @default.
• W3169063507 hasConcept C78458016 @default.
• W3169063507 hasConcept C86803240 @default.
• W3169063507 hasConceptScore W3169063507C119857082 @default.
• W3169063507 hasConceptScore W3169063507C122770356 @default.
• W3169063507 hasConceptScore W3169063507C14036430 @default.
• W3169063507 hasConceptScore W3169063507C153180895 @default.
• W3169063507 hasConceptScore W3169063507C154945302 @default.
• W3169063507 hasConceptScore W3169063507C167966045 @default.
• W3169063507 hasConceptScore W3169063507C205711294 @default.
• W3169063507 hasConceptScore W3169063507C39890363 @default.
• W3169063507 hasConceptScore W3169063507C41008148 @default.
• W3169063507 hasConceptScore W3169063507C51167844 @default.
• W3169063507 hasConceptScore W3169063507C73555534 @default.
• W3169063507 hasConceptScore W3169063507C78458016 @default.
• W3169063507 hasConceptScore W3169063507C86803240 @default.
• W3169063507 hasLocation W31690635071 @default.
• W3169063507 hasOpenAccess W3169063507 @default.
• W3169063507 hasPrimaryLocation W31690635071 @default.
• W3169063507 hasRelatedWork W1520325657 @default.
• W3169063507 hasRelatedWork W2116658704 @default.
• W3169063507 hasRelatedWork W2165279024 @default.
• W3169063507 hasRelatedWork W2247318628 @default.
• W3169063507 hasRelatedWork W2798914089 @default.

• next