Matches in SemOpenAlex for { <https://semopenalex.org/work/W3038325879> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3038325879 abstract "Recent studies show that deep neural networks are vulnerable to adversarial examples which can be generated via certain types of transformations. Being robust to a desired family of adversarial attacks is then equivalent to being invariant to a family of transformations. Learning invariant representations then naturally emerges as an important goal to achieve which we explore in this paper within specific application contexts. Specifically, we propose a cyclically-trained adversarial network to learn a mapping from image space to latent representation space and back such that the latent representation is invariant to a specified factor of variation (e.g., identity). The learned mapping assures that the synthesized image is not only realistic, but has the same values for unspecified factors (e.g., pose and illumination) as the original image and a desired value of the specified factor. Unlike disentangled representation learning, which requires two latent spaces, one for specified and another for unspecified factors, invariant representation learning needs only one such space. We encourage invariance to a specified factor by applying adversarial training using a variational autoencoder in the image space as opposed to the latent space. We strengthen this invariance by introducing a cyclic training process (forward and backward cycle). We also propose a new method to evaluate conditional generative networks. It compares how well different factors of variation can be predicted from the synthesized, as opposed to real, images. In quantitative terms, our approach attains state-of-the-art performance in experiments spanning three datasets with factors such as identity, pose, illumination or style. Our method produces sharp, high-quality synthetic images with little visible arte-facts compared to previous approaches." @default.
- W3038325879 created "2020-07-10" @default.
- W3038325879 creator A5028225823 @default.
- W3038325879 creator A5036913803 @default.
- W3038325879 creator A5063199641 @default.
- W3038325879 date "2020-06-01" @default.
- W3038325879 modified "2023-09-25" @default.
- W3038325879 title "A Cyclically-Trained Adversarial Network for Invariant Representation Learning" @default.
- W3038325879 cites W2010625607 @default.
- W3038325879 cites W2097246321 @default.
- W3038325879 cites W2124386111 @default.
- W3038325879 cites W2125874614 @default.
- W3038325879 cites W2163922914 @default.
- W3038325879 cites W2170653751 @default.
- W3038325879 cites W2344839403 @default.
- W3038325879 cites W2399452755 @default.
- W3038325879 cites W2512351403 @default.
- W3038325879 cites W2537024514 @default.
- W3038325879 cites W2737047298 @default.
- W3038325879 cites W2798434327 @default.
- W3038325879 cites W2799121188 @default.
- W3038325879 cites W2963278610 @default.
- W3038325879 cites W2963767194 @default.
- W3038325879 cites W2971626200 @default.
- W3038325879 doi "https://doi.org/10.1109/cvprw50498.2020.00399" @default.
- W3038325879 hasPublicationYear "2020" @default.
- W3038325879 type Work @default.
- W3038325879 sameAs 3038325879 @default.
- W3038325879 citedByCount "2" @default.
- W3038325879 countsByYear W30383258792023 @default.
- W3038325879 crossrefType "proceedings-article" @default.
- W3038325879 hasAuthorship W3038325879A5028225823 @default.
- W3038325879 hasAuthorship W3038325879A5036913803 @default.
- W3038325879 hasAuthorship W3038325879A5063199641 @default.
- W3038325879 hasBestOaLocation W30383258792 @default.
- W3038325879 hasConcept C101738243 @default.
- W3038325879 hasConcept C119857082 @default.
- W3038325879 hasConcept C121332964 @default.
- W3038325879 hasConcept C153180895 @default.
- W3038325879 hasConcept C154945302 @default.
- W3038325879 hasConcept C17744445 @default.
- W3038325879 hasConcept C190470478 @default.
- W3038325879 hasConcept C199539241 @default.
- W3038325879 hasConcept C24890656 @default.
- W3038325879 hasConcept C2776359362 @default.
- W3038325879 hasConcept C2778355321 @default.
- W3038325879 hasConcept C33923547 @default.
- W3038325879 hasConcept C37736160 @default.
- W3038325879 hasConcept C37914503 @default.
- W3038325879 hasConcept C41008148 @default.
- W3038325879 hasConcept C50644808 @default.
- W3038325879 hasConcept C59404180 @default.
- W3038325879 hasConcept C80444323 @default.
- W3038325879 hasConcept C94625758 @default.
- W3038325879 hasConceptScore W3038325879C101738243 @default.
- W3038325879 hasConceptScore W3038325879C119857082 @default.
- W3038325879 hasConceptScore W3038325879C121332964 @default.
- W3038325879 hasConceptScore W3038325879C153180895 @default.
- W3038325879 hasConceptScore W3038325879C154945302 @default.
- W3038325879 hasConceptScore W3038325879C17744445 @default.
- W3038325879 hasConceptScore W3038325879C190470478 @default.
- W3038325879 hasConceptScore W3038325879C199539241 @default.
- W3038325879 hasConceptScore W3038325879C24890656 @default.
- W3038325879 hasConceptScore W3038325879C2776359362 @default.
- W3038325879 hasConceptScore W3038325879C2778355321 @default.
- W3038325879 hasConceptScore W3038325879C33923547 @default.
- W3038325879 hasConceptScore W3038325879C37736160 @default.
- W3038325879 hasConceptScore W3038325879C37914503 @default.
- W3038325879 hasConceptScore W3038325879C41008148 @default.
- W3038325879 hasConceptScore W3038325879C50644808 @default.
- W3038325879 hasConceptScore W3038325879C59404180 @default.
- W3038325879 hasConceptScore W3038325879C80444323 @default.
- W3038325879 hasConceptScore W3038325879C94625758 @default.
- W3038325879 hasLocation W30383258791 @default.
- W3038325879 hasLocation W30383258792 @default.
- W3038325879 hasOpenAccess W3038325879 @default.
- W3038325879 hasPrimaryLocation W30383258791 @default.
- W3038325879 hasRelatedWork W2292254049 @default.
- W3038325879 hasRelatedWork W2592385986 @default.
- W3038325879 hasRelatedWork W2897995864 @default.
- W3038325879 hasRelatedWork W2937381246 @default.
- W3038325879 hasRelatedWork W2998168123 @default.
- W3038325879 hasRelatedWork W3099179464 @default.
- W3038325879 hasRelatedWork W3199009818 @default.
- W3038325879 hasRelatedWork W4281924768 @default.
- W3038325879 hasRelatedWork W4287995534 @default.
- W3038325879 hasRelatedWork W4296155043 @default.
- W3038325879 isParatext "false" @default.
- W3038325879 isRetracted "false" @default.
- W3038325879 magId "3038325879" @default.
- W3038325879 workType "article" @default.