Unsupervised Domain Adaptation using Graph Transduction Games

Vascon, Sebastiano; Aslan, Sinem; Torcinovich, Alessandro; Laarhoven, Twan van; Marchiori, Elena; Pelillo, Marcello

doi:10.1109/ijcnn.2019.8852075

Cited by 8 publications

(5 citation statements)

References 25 publications

(37 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, Table 3 compares UFAL to SOTA methods on the Office-Caltech dataset. Again, we surpass recently proposed methods such as GTDA+LR [32] and RWA [31]. UFAL even outperforms the ensemble based algorithm of Rakshit et al [29] by 0.5%.…”

Section: Resultsmentioning

confidence: 54%

Unsupervised Domain Adaptation by Uncertain Feature Alignment

Ringwald,

Stiefelhagen

2020

Preprint

View full text Add to dashboard Cite

Unsupervised domain adaptation (UDA) deals with the adaptation of models from a given source domain with labeled data to an unlabeled target domain. In this paper, we utilize the inherent prediction uncertainty of a model to accomplish the domain adaptation task. The uncertainty is measured by Monte-Carlo dropout and used for our proposed Uncertainty-based Filtering and Feature Alignment (UFAL) that combines an Uncertain Feature Loss (UFL) function and an Uncertainty-Based Filtering (UBF) approach for alignment of features in Euclidean space. Our method surpasses recently proposed architectures and achieves state-of-the-art results on multiple challenging datasets. Code is available on the project website.

show abstract

Section: Resultsmentioning

confidence: 54%

Unsupervised Domain Adaptation by Uncertain Feature Alignment

Ringwald,

Stiefelhagen

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Furthermore, Zhang et al [6] and Wang et al [7] combined the techniques in both branches. Vascon et al [8] and Wulfmeier et al [9] suggested a new technique for the UDA problem using Nash equilibrium [10] and Generative Adversarial Networks (GAN) [11], respectively.…”

Section: Introductionmentioning

confidence: 99%

Multi-Step Online Unsupervised Domain Adaptation

Moon

Das

Lee

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we address the Online Unsupervised Domain Adaptation (OUDA) problem, where the target data are unlabelled and arriving sequentially. The traditional methods on the OUDA problem mainly focus on transforming each arriving target data to the source domain, and they do not sufficiently consider the temporal coherency and accumulative statistics among the arriving target data. We propose a multi-step framework for the OUDA problem, which institutes a novel method to compute the mean-target subspace inspired by the geometrical interpretation on the Euclidean space. This meantarget subspace contains accumulative temporal information among the arrived target data. Moreover, the transformation matrix computed from the mean-target subspace is applied to the next target data as a preprocessing step, aligning the target data closer to the source domain. Experiments on four datasets demonstrated the contribution of each step in our proposed multi-step OUDA framework and its performance over previous approaches.Index Terms-Unsupervised domain adaptation, online domain adaptation, mean subspace, Grassmann manifold.

show abstract

“…The GTG has been succesfully applied in many different SSL contexts, like deep metric learning [10], matrix factorization [37], image recognition [2], proteinfunction prediction [39] and, indeed, traditional text-based WSD setting [36]. Moreover, it works consistently better [38] than other graph-based SSL methods like Label Propagation [41] and Gaussian Fields [42].…”

Section: Introductionmentioning

confidence: 99%

Transductive Visual Verb Sense Disambiguation

Vascon,

Aslan,

Bigaglia

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Verb Sense Disambiguation is a well-known task in NLP, the aim is to find the correct sense of a verb in a sentence. Recently, this problem has been extended in a multimodal scenario, by exploiting both textual and visual features of ambiguous verbs leading to a new problem, the Visual Verb Sense Disambiguation (VVSD). Here, the sense of a verb is assigned considering the content of an image paired with it rather than a sentence in which the verb appears. Annotating a dataset for this task is more complex than textual disambiguation, because assigning the correct sense to a pair of requires both non-trivial linguistic and visual skills. In this work, differently from the literature, the VVSD task will be performed in a transductive semi-supervised learning (SSL) setting, in which only a small amount of labeled information is required, reducing tremendously the need for annotated data. The disambiguation process is based on a graph-based label propagation method which takes into account mono or multimodal representations for pairs. Experiments have been carried out on the recently published dataset VerSe, the only available dataset for this task. The achieved results outperform the current state-of-the-art by a large margin while using only a small fraction of labeled samples per sense 1 .

show abstract

Unsupervised Domain Adaptation using Graph Transduction Games

Cited by 8 publications

References 25 publications

Unsupervised Domain Adaptation by Uncertain Feature Alignment

Unsupervised Domain Adaptation by Uncertain Feature Alignment

Multi-Step Online Unsupervised Domain Adaptation

Transductive Visual Verb Sense Disambiguation

Contact Info

Product

Resources

About