Matches in SemOpenAlex for { <https://semopenalex.org/work/W3086142364> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3086142364 endingPage "3331" @default.
- W3086142364 startingPage "3318" @default.
- W3086142364 abstract "Class-conditional generative models have gained popularity due to their characteristics of learning disentangled representations. However, these models typically require labeled examples in training. In this paper, we explore the feasibility of training these models on completely unlabeled data, under the assumption that we have access to other labeled data. The labeled data share the same label space, while their domain is shifted. Our model, which we refer to as KTransGAN, incorporates a classifier to transfer knowledge from the labeled data and performs collaborative learning with the conditional generator. By adopting these measures, KTransGAN is able to approximate the conditional distribution of the unlabeled data and simultaneously introduces a new solution to the unsupervised domain adaptation problem. To mitigate the training difficulty of our generative adversarial networks-based model, variational encoding and feature matching are also considered. From the empirical results, KTransGAN exhibits outstanding performance on a number of synthetic datasets and multiple real-world benchmarks. The quality of the synthesized instances is far superior to the pure variational autoencoding model. For example, on the CIFAR-10 dataset, our model scores 35.3 in FID, while the other model scores 128.45. In addition, the synthesis quality is close to the case when the model is trained in a fully supervised setting over the same number of training iterations. Regarding the classification performance, for instance, our model surpasses the highest state-of-the-art results (89.19%) by a large margin and achieves a test accuracy of 95.31% on the unlabeled data SVHN, while MNIST represents the labeled data. These results highlight the effectiveness of our proposed framework." @default.
- W3086142364 created "2020-09-21" @default.
- W3086142364 creator A5012772285 @default.
- W3086142364 creator A5014196059 @default.
- W3086142364 creator A5015819673 @default.
- W3086142364 creator A5040726112 @default.
- W3086142364 creator A5070982537 @default.
- W3086142364 date "2021-01-01" @default.
- W3086142364 modified "2023-10-14" @default.
- W3086142364 title "KTransGAN: Variational Inference-Based Knowledge Transfer for Unsupervised Conditional Generative Learning" @default.
- W3086142364 cites W2104094955 @default.
- W3086142364 cites W2194775991 @default.
- W3086142364 cites W2584009249 @default.
- W3086142364 cites W2593768305 @default.
- W3086142364 cites W2605488490 @default.
- W3086142364 cites W2617027347 @default.
- W3086142364 cites W2799907879 @default.
- W3086142364 cites W2885018852 @default.
- W3086142364 cites W2912423076 @default.
- W3086142364 cites W2919206487 @default.
- W3086142364 cites W2944388686 @default.
- W3086142364 cites W2962687275 @default.
- W3086142364 cites W2962770929 @default.
- W3086142364 cites W2962793481 @default.
- W3086142364 cites W2962852342 @default.
- W3086142364 cites W2963037989 @default.
- W3086142364 cites W2963150697 @default.
- W3086142364 cites W2963426391 @default.
- W3086142364 cites W2964159205 @default.
- W3086142364 cites W2982259084 @default.
- W3086142364 doi "https://doi.org/10.1109/tmm.2020.3023792" @default.
- W3086142364 hasPublicationYear "2021" @default.
- W3086142364 type Work @default.
- W3086142364 sameAs 3086142364 @default.
- W3086142364 citedByCount "6" @default.
- W3086142364 countsByYear W30861423642021 @default.
- W3086142364 countsByYear W30861423642022 @default.
- W3086142364 countsByYear W30861423642023 @default.
- W3086142364 crossrefType "journal-article" @default.
- W3086142364 hasAuthorship W3086142364A5012772285 @default.
- W3086142364 hasAuthorship W3086142364A5014196059 @default.
- W3086142364 hasAuthorship W3086142364A5015819673 @default.
- W3086142364 hasAuthorship W3086142364A5040726112 @default.
- W3086142364 hasAuthorship W3086142364A5070982537 @default.
- W3086142364 hasConcept C119857082 @default.
- W3086142364 hasConcept C150899416 @default.
- W3086142364 hasConcept C153180895 @default.
- W3086142364 hasConcept C154945302 @default.
- W3086142364 hasConcept C167966045 @default.
- W3086142364 hasConcept C190502265 @default.
- W3086142364 hasConcept C2776214188 @default.
- W3086142364 hasConcept C39890363 @default.
- W3086142364 hasConcept C41008148 @default.
- W3086142364 hasConcept C50644808 @default.
- W3086142364 hasConcept C774472 @default.
- W3086142364 hasConcept C95623464 @default.
- W3086142364 hasConceptScore W3086142364C119857082 @default.
- W3086142364 hasConceptScore W3086142364C150899416 @default.
- W3086142364 hasConceptScore W3086142364C153180895 @default.
- W3086142364 hasConceptScore W3086142364C154945302 @default.
- W3086142364 hasConceptScore W3086142364C167966045 @default.
- W3086142364 hasConceptScore W3086142364C190502265 @default.
- W3086142364 hasConceptScore W3086142364C2776214188 @default.
- W3086142364 hasConceptScore W3086142364C39890363 @default.
- W3086142364 hasConceptScore W3086142364C41008148 @default.
- W3086142364 hasConceptScore W3086142364C50644808 @default.
- W3086142364 hasConceptScore W3086142364C774472 @default.
- W3086142364 hasConceptScore W3086142364C95623464 @default.
- W3086142364 hasFunder F4320309893 @default.
- W3086142364 hasFunder F4320321001 @default.
- W3086142364 hasFunder F4320321921 @default.
- W3086142364 hasFunder F4320335787 @default.
- W3086142364 hasLocation W30861423641 @default.
- W3086142364 hasOpenAccess W3086142364 @default.
- W3086142364 hasPrimaryLocation W30861423641 @default.
- W3086142364 hasRelatedWork W2885835503 @default.
- W3086142364 hasRelatedWork W2906272760 @default.
- W3086142364 hasRelatedWork W2950475743 @default.
- W3086142364 hasRelatedWork W2978098801 @default.
- W3086142364 hasRelatedWork W2979740303 @default.
- W3086142364 hasRelatedWork W3104832546 @default.
- W3086142364 hasRelatedWork W3160820590 @default.
- W3086142364 hasRelatedWork W4289105138 @default.
- W3086142364 hasRelatedWork W4310699748 @default.
- W3086142364 hasRelatedWork W4386603768 @default.
- W3086142364 hasVolume "23" @default.
- W3086142364 isParatext "false" @default.
- W3086142364 isRetracted "false" @default.
- W3086142364 magId "3086142364" @default.
- W3086142364 workType "article" @default.