litlas
ログインアカウント作成

Semi-Supervised Optimal Transport for Heterogeneous Domain Adaptation

著者: Yuguang Yan, Wen Li, Hanrui Wu, Huaqing Min, Mingkui Tan, Qingyao Wu - International Joint Conference on Artificial Intelligence, IJCAI 2018 被引用: 118

Heterogeneous domain adaptation (HDA) aims to exploit knowledge from a heterogeneous source domain to improve the learning performance in a target domain. Since the feature spaces of the source and target domains are different, the transferring of knowledge is extremely difficult. In this paper, we propose a novel semi-supervised algorithm for HDA by exploiting the theory of optimal transport (OT), a powerful tool originally designed for aligning two different distributions. To match the samples between heterogeneous domains, we propose to preserve the semantic consistency between heterogeneous domains by incorporating label information into the entropic Gromov-Wasserstein discrepancy, which is a metric in OT for different metric spaces, resulting in a new semi-supervised scheme. Via the new scheme, the target and transported source samples with the same label are enforced to follow similar distributions. Lastly, based on the Kullback-Leibler metric, we develop an efficient algorithm to optimize the resultant problem. Comprehensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of our proposed method.

Learning with AugmentedFeatures for…Learning with Augmented Features for Heterogeneous Domain AdaptationLearning With AugmentedFeatures for Supervised…Learning With Augmented Features for Supervised and Semi-Supervised Heterogeneous Domain AdaptationHybrid HeterogeneousTransfer Learning…Hybrid Heterogeneous Transfer Learning through Deep LearningOptimal Transport forDomain AdaptationOptimal Transport for Domain AdaptationLearning TransferableFeatures with Deep…Learning Transferable Features with Deep Adaptation NetworksLearning Cross-DomainLandmarks for…Learning Cross-Domain Landmarks for Heterogeneous Domain AdaptationTransfer Learning forCross-Language Text…Transfer Learning for Cross-Language Text Categorization through Active Correspondences ConstructionGromov-WassersteinAveraging of Kernel and…Gromov-Wasserstein Averaging of Kernel and Distance MatricesMapping Estimation forDiscrete Optimal…Mapping Estimation for Discrete Optimal TransportLearning DiscriminativeCorrelation Subspace fo…Learning Discriminative Correlation Subspace for Heterogeneous Domain AdaptationOnline HeterogeneousTransfer by Hedge…Online Heterogeneous Transfer by Hedge Ensemble of Offline and Online DecisionsCompletely HeterogeneousTransfer Learning with…Completely Heterogeneous Transfer Learning with Attention - What And What Not To TransferHeterogeneous TransferLearning via Deep Matri…Heterogeneous Transfer Learning via Deep Matrix Completion with Adversarial Kernel EmbeddingGeometric KnowledgeEmbedding for…Geometric Knowledge Embedding for unsupervised domain adaptationLearning GenerativeModels across…Learning Generative Models across Incomparable SpacesInformative FeatureSelection for Domain…Informative Feature Selection for Domain AdaptationHeterogeneous DomainAdaptation via Soft…Heterogeneous Domain Adaptation via Soft Transfer NetworkIterative Refinement forMulti-Source Visual…Iterative Refinement for Multi-Source Visual Domain AdaptationDiscriminativedistribution alignment…Discriminative distribution alignment: A unified framework for heterogeneous domain adaptationHeterogeneous DomainAdaptation by…Heterogeneous Domain Adaptation by Information Capturing and Distribution MatchingTransferable FeatureSelection for…Transferable Feature Selection for Unsupervised Domain AdaptationKnowledge Preserving andDistribution Alignment…Knowledge Preserving and Distribution Alignment for Heterogeneous Domain AdaptationMultisourceHeterogeneous Domain…Multisource Heterogeneous Domain Adaptation With Conditional Weighting Adversarial NetworkGraph embedding-basedheterogeneous domain…Graph embedding-based heterogeneous domain adaptation with domain-invariant feature learning and distributional order preservingSemi-Supervised OptimalTransport for…Semi-Supervised Optimal Transport for Heterogeneous Domain Adaptation過去の参考文献中心の論文この論文を引用する論文古い新しい

ノードをクリックするとフォーカスを固定、空白をクリックすると本論文に戻ります。ホバーで一時的にプレビューできます。各ノードのページはタイトルから開けます。