Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136392033> ?p ?o ?g. }
- W3136392033 endingPage "44002" @default.
- W3136392033 startingPage "43991" @default.
- W3136392033 abstract "Autoencoder based methods are the majority of deep unsupervised outlier detection methods. However, these methods perform not well on complex image datasets and suffer from the noise introduced by outliers, especially when the outlier ratio is high. In this paper, we propose a framework named Transformation Invariant AutoEncoder (TIAE), which can achieve stable and high performance on unsupervised outlier detection. First, instead of using a conventional autoencoder, we propose a transformation invariant autoencoder to do better representation learning for complex image datasets. Next, to mitigate the negative effect of noise introduced by outliers and stabilize the network training, we select the most confident inliers likely examples in each epoch as the training set by incorporating adaptive self-paced learning in our TIAE framework. Extensive evaluations show that TIAE significantly advances unsupervised outlier detection performance by up to 10% AUROC against other autoencoder based methods on five image datasets." @default.
- W3136392033 created "2021-03-29" @default.
- W3136392033 creator A5025331095 @default.
- W3136392033 creator A5040027560 @default.
- W3136392033 creator A5059620114 @default.
- W3136392033 creator A5064317694 @default.
- W3136392033 creator A5069681054 @default.
- W3136392033 date "2021-01-01" @default.
- W3136392033 modified "2023-10-12" @default.
- W3136392033 title "Unsupervised Outlier Detection via Transformation Invariant Autoencoder" @default.
- W3136392033 cites W1974879849 @default.
- W3136392033 cites W1976526581 @default.
- W3136392033 cites W2049058890 @default.
- W3136392033 cites W2057245489 @default.
- W3136392033 cites W2084512860 @default.
- W3136392033 cites W2089468765 @default.
- W3136392033 cites W2112796928 @default.
- W3136392033 cites W2127979711 @default.
- W3136392033 cites W2136655611 @default.
- W3136392033 cites W2204904589 @default.
- W3136392033 cites W2296719434 @default.
- W3136392033 cites W2342408547 @default.
- W3136392033 cites W2358876993 @default.
- W3136392033 cites W2740656851 @default.
- W3136392033 cites W2743138268 @default.
- W3136392033 cites W2747648188 @default.
- W3136392033 cites W2779692282 @default.
- W3136392033 cites W2889081631 @default.
- W3136392033 cites W2901412525 @default.
- W3136392033 cites W2911867527 @default.
- W3136392033 cites W2938414802 @default.
- W3136392033 cites W2939176882 @default.
- W3136392033 cites W2948975938 @default.
- W3136392033 cites W2963061824 @default.
- W3136392033 cites W2963073614 @default.
- W3136392033 cites W2966559104 @default.
- W3136392033 cites W2966661 @default.
- W3136392033 cites W2978971541 @default.
- W3136392033 cites W2981747647 @default.
- W3136392033 cites W2987228832 @default.
- W3136392033 cites W2991966331 @default.
- W3136392033 cites W2993502289 @default.
- W3136392033 cites W3034292309 @default.
- W3136392033 cites W3035524453 @default.
- W3136392033 cites W3080722105 @default.
- W3136392033 cites W3081175302 @default.
- W3136392033 cites W3110145754 @default.
- W3136392033 cites W3114010851 @default.
- W3136392033 cites W4239510810 @default.
- W3136392033 doi "https://doi.org/10.1109/access.2021.3065838" @default.
- W3136392033 hasPublicationYear "2021" @default.
- W3136392033 type Work @default.
- W3136392033 sameAs 3136392033 @default.
- W3136392033 citedByCount "6" @default.
- W3136392033 countsByYear W31363920332021 @default.
- W3136392033 countsByYear W31363920332022 @default.
- W3136392033 crossrefType "journal-article" @default.
- W3136392033 hasAuthorship W3136392033A5025331095 @default.
- W3136392033 hasAuthorship W3136392033A5040027560 @default.
- W3136392033 hasAuthorship W3136392033A5059620114 @default.
- W3136392033 hasAuthorship W3136392033A5064317694 @default.
- W3136392033 hasAuthorship W3136392033A5069681054 @default.
- W3136392033 hasBestOaLocation W31363920331 @default.
- W3136392033 hasConcept C101738243 @default.
- W3136392033 hasConcept C104317684 @default.
- W3136392033 hasConcept C108583219 @default.
- W3136392033 hasConcept C153180895 @default.
- W3136392033 hasConcept C154945302 @default.
- W3136392033 hasConcept C185592680 @default.
- W3136392033 hasConcept C190470478 @default.
- W3136392033 hasConcept C204241405 @default.
- W3136392033 hasConcept C33923547 @default.
- W3136392033 hasConcept C37914503 @default.
- W3136392033 hasConcept C41008148 @default.
- W3136392033 hasConcept C55493867 @default.
- W3136392033 hasConcept C59404180 @default.
- W3136392033 hasConcept C739882 @default.
- W3136392033 hasConcept C79337645 @default.
- W3136392033 hasConcept C8038995 @default.
- W3136392033 hasConceptScore W3136392033C101738243 @default.
- W3136392033 hasConceptScore W3136392033C104317684 @default.
- W3136392033 hasConceptScore W3136392033C108583219 @default.
- W3136392033 hasConceptScore W3136392033C153180895 @default.
- W3136392033 hasConceptScore W3136392033C154945302 @default.
- W3136392033 hasConceptScore W3136392033C185592680 @default.
- W3136392033 hasConceptScore W3136392033C190470478 @default.
- W3136392033 hasConceptScore W3136392033C204241405 @default.
- W3136392033 hasConceptScore W3136392033C33923547 @default.
- W3136392033 hasConceptScore W3136392033C37914503 @default.
- W3136392033 hasConceptScore W3136392033C41008148 @default.
- W3136392033 hasConceptScore W3136392033C55493867 @default.
- W3136392033 hasConceptScore W3136392033C59404180 @default.
- W3136392033 hasConceptScore W3136392033C739882 @default.
- W3136392033 hasConceptScore W3136392033C79337645 @default.
- W3136392033 hasConceptScore W3136392033C8038995 @default.
- W3136392033 hasFunder F4320321001 @default.
- W3136392033 hasFunder F4320322843 @default.
- W3136392033 hasFunder F4320324150 @default.