Abstract:We address the problem of web supervised learning, in particular for face attribute classification. Web data suffers from image set noise, due to unrelated images that may be retrieved in response to the query. We propose a semi-supervised pseudo-labeling approach where the embedding space distribution is learnt via variational contrastive learning. We use 40 Gaussian sampling heads for the 40 attributes in the CelebA dataset and apply supervised contrastive learning over a limited amount of labelled data, to … Show more
“…Advanced autoencoder deep learning models make extracting information from heterogeneous contexts more efficient [6]. In web image search, semi-supervised pseudo-labeling and variational contrastive learning can be used to overcome the influence of noise and obtain better retrieval performance [7]. Embracing location-based social networks into web applications enables the users to register whenever they visit a specific point-of-interest (POI) through the so-called check-ins, or to establish social links with other users in the system [8].…”
In today’s era of rapid digitization and information technology advancement, web search and web data mining stand at the core of the technological progress of numerous web-based applications [...]
“…Advanced autoencoder deep learning models make extracting information from heterogeneous contexts more efficient [6]. In web image search, semi-supervised pseudo-labeling and variational contrastive learning can be used to overcome the influence of noise and obtain better retrieval performance [7]. Embracing location-based social networks into web applications enables the users to register whenever they visit a specific point-of-interest (POI) through the so-called check-ins, or to establish social links with other users in the system [8].…”
In today’s era of rapid digitization and information technology advancement, web search and web data mining stand at the core of the technological progress of numerous web-based applications [...]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.