Multi-domain Adversarial Variational Bayesian Inference for Domain Generalization
dc.contributor.author | Gao, Zhifan | |
dc.contributor.author | Guo, Saidi | |
dc.contributor.author | Xu, Chenchu | |
dc.contributor.author | Zhang, Jinglin | |
dc.contributor.author | Gong, Mingming | |
dc.contributor.author | Del Ser, Javier | |
dc.contributor.author | Li, Shuo | |
dc.contributor.institution | IA | |
dc.date.accessioned | 2024-07-24T12:11:19Z | |
dc.date.available | 2024-07-24T12:11:19Z | |
dc.date.issued | 2022 | |
dc.description | Publisher Copyright: IEEE | |
dc.description.abstract | Domain generalization aims to learn common knowledge from multiple observed source domains and transfer it to unseen target domains, e.g. the object recognition in varieties of visual environments. Traditional domain generalization methods aim to learn the feature representation of the raw data with its distribution invariant across domains. This relies on the assumption that the two posterior distributions (the distributions of the label given the feature distribution and given the raw data) are stable in different domains. However, this does not always hold in many practical situations. In this paper, we relax the above assumption by permitting the posterior distribution of the label given the raw data changes in difference domains, and thus focuses on a more realistic learning problem that infers the conditional domain-invariant feature representation. Specifically, a multi-domain adversarial variational Bayesian inference approach is proposed to minimize the inter-domain discrepancy of the conditional distributions of the feature given the label. Besides, it is imposed by the constraints from the adversarial learning and feedback mechanism to enhance the condition invariant feature representation. The extensive experiments on two datasets demonstrate the effectiveness of our approach, as well as the state-of-the-art performance comparing with thirteen methods. | en |
dc.description.status | Peer reviewed | |
dc.format.extent | 1 | |
dc.identifier.citation | Gao , Z , Guo , S , Xu , C , Zhang , J , Gong , M , Del Ser , J & Li , S 2022 , ' Multi-domain Adversarial Variational Bayesian Inference for Domain Generalization ' , IEEE Transactions on Circuits and Systems for Video Technology , pp. 1 . https://doi.org/10.1109/TCSVT.2022.3232112 | |
dc.identifier.doi | 10.1109/TCSVT.2022.3232112 | |
dc.identifier.issn | 1051-8215 | |
dc.identifier.uri | https://hdl.handle.net/11556/4171 | |
dc.identifier.url | http://www.scopus.com/inward/record.url?scp=85146234522&partnerID=8YFLogxK | |
dc.language.iso | eng | |
dc.relation.ispartof | IEEE Transactions on Circuits and Systems for Video Technology | |
dc.rights | info:eu-repo/semantics/restrictedAccess | |
dc.subject.keywords | Adaptation models | |
dc.subject.keywords | Adversarial machine learning | |
dc.subject.keywords | Bayes methods | |
dc.subject.keywords | Biomedical imaging | |
dc.subject.keywords | Domain generalization | |
dc.subject.keywords | Task analysis | |
dc.subject.keywords | Training | |
dc.subject.keywords | Visualization | |
dc.subject.keywords | variational auto-encoder | |
dc.subject.keywords | visual object recognition | |
dc.subject.keywords | Media Technology | |
dc.subject.keywords | Electrical and Electronic Engineering | |
dc.title | Multi-domain Adversarial Variational Bayesian Inference for Domain Generalization | en |
dc.type | journal article |