Matches in SemOpenAlex for { <https://semopenalex.org/work/W2928477066> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W2928477066 abstract "An autoencoder is a neural network which data projects to and from a lower dimensional latent space, where this data is easier to understand and model. The autoencoder consists of two sub-networks, the encoder and the decoder, which carry out these transformations. The neural network is trained such that the output is as close to the input as possible, the data having gone through an information bottleneck : the latent space. This tool bears significant ressemblance to Principal Component Analysis (PCA), with two main differences. Firstly, the autoencoder is a non-linear transformation, contrary to PCA, which makes the autoencoder more flexible and powerful. Secondly, the axes found by a PCA are orthogonal, and are ordered in terms of the amount of variability which the data presents along these axes. This makes the interpretability of the PCA much greater than that of the autoencoder, which does not have these attributes. Ideally, then, we would like an autoencoder whose latent space consists of independent components, ordered by decreasing importance to the data. In this paper, we propose an algorithm to create such a network. We create an iterative algorithm which progressively increases the size of the latent space, learning a new dimension at each step. Secondly, we propose a covariance loss term to add to the standard autoencoder loss function, as well as a normalisation layer just before the latent space, which encourages the latent space components to be statistically independent. We demonstrate the results of this autoencoder on simple geometric shapes, and find that the algorithm indeed finds a meaningful representation in the latent space. This means that subsequent interpolation in the latent space has meaning with respect to the geometric properties of the images." @default.
- W2928477066 created "2019-04-11" @default.
- W2928477066 creator A5032799266 @default.
- W2928477066 creator A5037770503 @default.
- W2928477066 creator A5047314512 @default.
- W2928477066 date "2019-04-02" @default.
- W2928477066 modified "2023-10-01" @default.
- W2928477066 title "A PCA-like Autoencoder." @default.
- W2928477066 cites W1522301498 @default.
- W2928477066 cites W1853900790 @default.
- W2928477066 cites W1959608418 @default.
- W2928477066 cites W2017257315 @default.
- W2928477066 cites W2099471712 @default.
- W2928477066 cites W2108665656 @default.
- W2928477066 cites W2156387975 @default.
- W2928477066 cites W2519536754 @default.
- W2928477066 cites W2606748186 @default.
- W2928477066 cites W2904367110 @default.
- W2928477066 cites W2949117887 @default.
- W2928477066 cites W2962752582 @default.
- W2928477066 cites W2962793481 @default.
- W2928477066 hasPublicationYear "2019" @default.
- W2928477066 type Work @default.
- W2928477066 sameAs 2928477066 @default.
- W2928477066 citedByCount "3" @default.
- W2928477066 countsByYear W29284770662019 @default.
- W2928477066 countsByYear W29284770662021 @default.
- W2928477066 crossrefType "posted-content" @default.
- W2928477066 hasAuthorship W2928477066A5032799266 @default.
- W2928477066 hasAuthorship W2928477066A5037770503 @default.
- W2928477066 hasAuthorship W2928477066A5047314512 @default.
- W2928477066 hasConcept C101738243 @default.
- W2928477066 hasConcept C111919701 @default.
- W2928477066 hasConcept C11413529 @default.
- W2928477066 hasConcept C118505674 @default.
- W2928477066 hasConcept C153180895 @default.
- W2928477066 hasConcept C154945302 @default.
- W2928477066 hasConcept C202444582 @default.
- W2928477066 hasConcept C27438332 @default.
- W2928477066 hasConcept C2778572836 @default.
- W2928477066 hasConcept C33676613 @default.
- W2928477066 hasConcept C33923547 @default.
- W2928477066 hasConcept C41008148 @default.
- W2928477066 hasConcept C50644808 @default.
- W2928477066 hasConcept C70518039 @default.
- W2928477066 hasConceptScore W2928477066C101738243 @default.
- W2928477066 hasConceptScore W2928477066C111919701 @default.
- W2928477066 hasConceptScore W2928477066C11413529 @default.
- W2928477066 hasConceptScore W2928477066C118505674 @default.
- W2928477066 hasConceptScore W2928477066C153180895 @default.
- W2928477066 hasConceptScore W2928477066C154945302 @default.
- W2928477066 hasConceptScore W2928477066C202444582 @default.
- W2928477066 hasConceptScore W2928477066C27438332 @default.
- W2928477066 hasConceptScore W2928477066C2778572836 @default.
- W2928477066 hasConceptScore W2928477066C33676613 @default.
- W2928477066 hasConceptScore W2928477066C33923547 @default.
- W2928477066 hasConceptScore W2928477066C41008148 @default.
- W2928477066 hasConceptScore W2928477066C50644808 @default.
- W2928477066 hasConceptScore W2928477066C70518039 @default.
- W2928477066 hasLocation W29284770661 @default.
- W2928477066 hasOpenAccess W2928477066 @default.
- W2928477066 hasPrimaryLocation W29284770661 @default.
- W2928477066 hasRelatedWork W1771781207 @default.
- W2928477066 hasRelatedWork W2027027919 @default.
- W2928477066 hasRelatedWork W2100495367 @default.
- W2928477066 hasRelatedWork W2112796928 @default.
- W2928477066 hasRelatedWork W2126461605 @default.
- W2928477066 hasRelatedWork W2344593201 @default.
- W2928477066 hasRelatedWork W2754768167 @default.
- W2928477066 hasRelatedWork W2884797809 @default.
- W2928477066 hasRelatedWork W2886752954 @default.
- W2928477066 hasRelatedWork W2913796826 @default.
- W2928477066 hasRelatedWork W2919474366 @default.
- W2928477066 hasRelatedWork W2951311921 @default.
- W2928477066 hasRelatedWork W2990902486 @default.
- W2928477066 hasRelatedWork W2997952090 @default.
- W2928477066 hasRelatedWork W3035042347 @default.
- W2928477066 hasRelatedWork W3103473572 @default.
- W2928477066 hasRelatedWork W3103988127 @default.
- W2928477066 hasRelatedWork W3119769898 @default.
- W2928477066 hasRelatedWork W3130797375 @default.
- W2928477066 hasRelatedWork W3196858806 @default.
- W2928477066 isParatext "false" @default.
- W2928477066 isRetracted "false" @default.
- W2928477066 magId "2928477066" @default.
- W2928477066 workType "article" @default.