Cleaning Noisy Labels by Negative Ensemble Learning for Source-Free Unsupervised Domain Adaptation

Ahmed, Waqar; Morerio, Pietro; Murino, Vittorio

doi:10.1109/wacv51458.2022.00043

Cited by 30 publications

(12 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Source-free Domain Adaptation (SFDA) aims to adapt a model to the unlabelled target domain when only the source model is available and the source data set is absent during target adaptation. Existing SFDA methods use pseudolabel refinement [35,1,5], latent source feature generation using variational inference [74], or disparity among an ensemble of classifiers [31]. Certain SFDA methods resort to ad hoc source training protocols to enable the source model to be adapted on the target data.…”

Section: Related Workmentioning

confidence: 99%

“…On the other hand, our proposed U-SFAN does not require specialized source training except a computationally lightweight approximate inference, which can be done with a single pass of the source data during the source training. Moreover, unlike [31,1], our U-SFAN works well on both closed-set and open-set SFDA without ad hoc modifications.…”

Section: Related Workmentioning

confidence: 99%

“…ResNet-10155 1. 53.3 61.9 59.1 80.6 17.9 79.7 31.2 81.0 26.5 73.5 8.5 52.4 DANN [15] 81.9 77.7 82.8 44.3 81.2 29.5 65.1 28.6 51.9 54.6 82.8 7.8 57.4 ADR [59] 94.2 48.5 84.0 72.9 90.1 74.2 92.6 72.5 80.8 61.8 82.2 28.8 73.5 CDAN [40] 85.2 66.9 83.0 50.8 84.2 74.9 88.1 74.5 83.4 76.0 81.9 38.0 73.9 CDAN+BSP [6] 92.4 61.0 81.0 57.5 89.0 80.6 90.1 77.0 84.2 77.9 82.1 38.4 75.9 .4 68.8 74.8 93.1 95.4 88.6 84.7 89.1 84.7 83.5 48.1 81.6 A 2 Net[71] 94.0 87.8 85.6 66.8 93.7 95.1 85.8 81.2 91.6 88.2 86.5 56.0 84.3 SHOT † [35] 94.9 87.1 76.9 55.0 94.2 95.4 80.8 80.0 89.5 88.7 85.6 60.5 82.4 U-SFAN + (Ours) 94.9 87.4 78.0 56.4 93.8 95.1 80.5 79.9 90.1 90.1 85.3 60.4 82.7…”

mentioning

confidence: 99%

See 2 more Smart Citations

Uncertainty-Guided Source-Free Domain Adaptation

Roy

Trapp

Pilzer

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Source-free domain adaptation (SFDA) aims to adapt a classifier to an unlabelled target data set by only using a pre-trained source model. However, the absence of the source data and the domain shift makes the predictions on the target data unreliable. We propose quantifying the uncertainty in the source model predictions and utilizing it to guide the target adaptation. For this, we construct a probabilistic source model by incorporating priors on the network parameters inducing a distribution over the model predictions. Uncertainties are estimated by employing a Laplace approximation and incorporated to identify target data points that do not lie in the source manifold and to down-weight them when maximizing the mutual information on the target data. Unlike recent works, our probabilistic treatment is computationally lightweight, decouples source training and target adaptation, and requires no specialized source training or changes of the model architecture. We show the advantages of uncertainty-guided SFDA over traditional SFDA in the closed-set and open-set settings and provide empirical evidence that our approach is more robust to strong domain shifts even without tuning.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Uncertainty-Guided Source-Free Domain Adaptation

Roy

Trapp

Pilzer

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…where c k denotes the centroid of k-th class, δ k denotes the k-th element in the soft-max operation. Some recent works [51], [94] argue that using only one prototype does not fully characterize the class, so multiple prototypes are generated in each class. In general, the aim of this step is to attain a class prototype p k that can represent the distribution of each category in the target domain.…”

Section: Pseudo Labelingmentioning

confidence: 99%

“…5. Therefore, the false label filtering [56], [87], [100] and the noisy label learning [94], [101], [102] are two important directions to improve the accuracy of pseudo labels. False label filtering is mainly to reject unreliable pseudo-labels by designing some reasonable mechanism.…”

Section: Pseudo Labelingmentioning

confidence: 99%

Source-Free Domain Adaptation for Real-World Image Dehazing

Huang

Liu

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Over the past decade, domain adaptation has become a widely studied branch of transfer learning that aims to improve performance on target domains by leveraging knowledge from the source domain. Conventional domain adaptation methods often assume access to both source and target domain data simultaneously, which may not be feasible in real-world scenarios due to privacy and confidentiality concerns. As a result, the research of Source-Free Domain Adaptation (SFDA) has drawn growing attention in recent years, which only utilizes the source-trained model and unlabeled target data to adapt to the target domain. Despite the rapid explosion of SFDA work, yet there has no timely and comprehensive survey in the field. To fill this gap, we provide a comprehensive survey of recent advances in SFDA and organize them into a unified categorization scheme based on the framework of transfer learning. Instead of presenting each approach independently, we modularize several components of each method to more clearly illustrate their relationships and mechanics in light of the composite properties of each method. Furthermore, we compare the results of more than 30 representative SFDA methods on three popular classification benchmarks, namely Office-31, Office-home, and VisDA, to explore the effectiveness of various technical routes and the combination effects among them. Additionally, we briefly introduce the applications of SFDA and related fields. Drawing from our analysis of the challenges facing SFDA, we offer some insights into future research directions and potential settings.

show abstract