Instance-Dependent Noisy Label Learning via Graphical Modelling

Simple item page
dc.contributor.authorGarg, A.
dc.contributor.authorNguyen, C.
dc.contributor.authorFelix, R.
dc.contributor.authorDo, T.-T.
dc.contributor.authorCarneiro, G.
dc.contributor.conferenceIEEE/CVF Winter Conference on Applications of Computer Vision (WACV) (2 Jan 2023 - 7 Jan 2023 : Waikoloa, HI, USA)
dc.date.issued2023
dc.description.abstractNoisy labels are unavoidable yet troublesome in the ecosystem of deep learning because models can easily overfit them. There are many types of label noise, such as symmetric, asymmetric and instance-dependent noise (IDN), with IDN being the only type that depends on image information. Such dependence on image information makes IDN a critical type of label noise to study, given that labelling mistakes are caused in large part by insufficient or ambiguous information about the visual classes present in images. Aiming to provide an effective technique to address IDN, we present a new graphical modelling approach called InstanceGM, that combines discriminative and generative models. The main contributions of InstanceGM are: i) the use of the continuous Bernoulli distribution to train the generative model, offering significant training advantages, and ii) the exploration of a state-of-the-art noisy-label discriminative classifier to generate clean labels from instancedependent noisy-label samples. InstanceGM is competitive with current noisy-label learning approaches, particularly in IDN benchmarks using synthetic and real-world datasets, where our method shows better accuracy than the competitors in most experiments¹.
dc.description.statementofresponsibilityArpit Garg, Cuong Nguyen, Rafael Felix, Thanh-Toan Do, and Gustavo Carneiro
dc.identifier.citationProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), 2023, pp.2287-2297
dc.identifier.doi10.1109/wacv56688.2023.00232
dc.identifier.isbn9781665493475
dc.identifier.issn2472-6737
dc.identifier.issn2642-9381
dc.identifier.orcidGarg, A. [0000-0001-5886-5822]
dc.identifier.orcidNguyen, C. [0000-0003-2672-6291]
dc.identifier.orcidCarneiro, G. [0000-0002-5571-6220]
dc.identifier.urihttps://hdl.handle.net/2440/139216
dc.language.isoen
dc.publisherIEEE
dc.publisher.placeOnline
dc.relation.granthttp://purl.org/au-research/grants/arc/DP180103232
dc.relation.granthttp://purl.org/au-research/grants/arc/FT190100525
dc.relation.ispartofseriesIEEE Winter Conference on Applications of Computer Vision
dc.rights©2023 IEEE
dc.source.urihttps://ieeexplore.ieee.org/xpl/conhome/10030081/proceeding
dc.titleInstance-Dependent Noisy Label Learning via Graphical Modelling
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Australian Institute for Machine Learning publications
Computer Science publications