Adaptive View-Aligned and Feature Augmentation Network for Partially View-Aligned Clustering

Zhang, Xianchao; Chen, Mengyan; Mu, Jie; Zong, Linlin

doi:10.1007/978-3-031-33374-3_18

Cited by 2 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…What is to say, we have less information available to learn the alignment matrix, which leads to poorer learning performance (because we only use unpaired samples in the alignment phase). PVC [29], one of the prior works, shows a similar result with a high alignment ratio. Second, a lower alignment rate does not always lead to poorer clustering performance.…”

Section: G Alignment Performancementioning

confidence: 55%

“…• Partially view-aligned clustering (PVC) [29]: This method uses a differentiable surrogate of the Hungarian algorithm to learn the correspondence of the unaligned data in a latent space learned by a neural network • Multi-view Contrastive Learning with Noise-robust Loss (MvCLN) [30]: This method proposes a novel noise-robust contrastive loss to establish the view correspondence using contrastive learning 3) Multi-View Clustering Algorithm on Fully Unpaired Data:…”

Section: Compared Methodsmentioning

confidence: 99%

“…[60] construct different types of constraint matrices in optimization to learn a desirable correspondence. [29] design a differentiable alignment module and integrate it into a deep neural network to reconstruct the view alignment relationship. [30] proposes multi-view contrastive learning with a noise robust loss learning model such that paired samples are in the representation space close to each other, while the unpaired samples are away from each other.…”

Section: B Multi-view Clustering On Unpaired Datamentioning

confidence: 99%

“…Several approaches have been proposed to solve this problem, [29] uses a differentiable surrogate of the Hungarian algorithm to learn the correspondence of the unaligned data in a latent space learned by a neural network. [30] proposes a noise-robust contrastive loss to establish the view correspondence.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Continual Multi-view Clustering

Wan

Liu

Liang

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Multi-view clustering (MVC) has gained broad attention owing to its capacity to exploit consistent and complementary information across views. This paper focuses on a challenging issue in MVC called the incomplete continual data problem (ICDP). In specific, most existing algorithms assume that views are available in advance and overlook the scenarios where data observations of views are accumulated over time. Due to privacy considerations or memory limitations, previous views cannot be stored in these situations. Some works are proposed to handle it, but all fail to address incomplete views. Such an incomplete continual data problem (ICDP) in MVC is tough to solve since incomplete information with continual data increases the difficulty of extracting consistent and complementary knowledge among views. We propose Fast Continual Multi-View Clustering with Incomplete Views (FCMVC-IV) to address it. Specifically, it maintains a consensus coefficient matrix and updates knowledge with the incoming incomplete view rather than storing and recomputing all the data matrices. Considering that the views are incomplete, the newly collected view might contain samples that have yet to appear; two indicator matrices and a rotation matrix are developed to match matrices with different dimensions. Besides, we design a three-step iterative algorithm to solve the resultant problem in linear complexity with proven convergence. Comprehensive experiments on various datasets show the superiority of FCMVC-IV.

show abstract

Section: G Alignment Performancementioning

confidence: 55%

Section: Compared Methodsmentioning

confidence: 99%

Section: B Multi-view Clustering On Unpaired Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Continual Multi-view Clustering

Wan

Liu

Liang

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…They usually consist of 3 steps: 1) measuring the consistency of views; 2) identifying outliers as instances with significant inconsistencies across views; 3) removing outliers to construct a clean dataset. Recently, some multi-view learning methods (Huang et al 2020;Zhang et al 2023d) dedicate to learning alignment relations for the original data and constructing new data instances accordingly. For example, the text view in Fig.…”

Section: Introductionmentioning

confidence: 99%

Reliable Conflictive Multi-View Learning

Xu,

Si,

Guan

et al. 2024

AAAI

View full text Add to dashboard Cite

Multi-view learning aims to combine multiple features to achieve more comprehensive descriptions of data. Most previous works assume that multiple views are strictly aligned. However, real-world multi-view data may contain low-quality conflictive instances, which show conflictive information in different views. Previous methods for this problem mainly focus on eliminating the conflictive data instances by removing them or replacing conflictive views. Nevertheless, real-world applications usually require making decisions for conflictive instances rather than only eliminating them. To solve this, we point out a new Reliable Conflictive Multi-view Learning (RCML) problem, which requires the model to provide decision results and attached reliabilities for conflictive multi-view data. We develop an Evidential Conflictive Multi-view Learning (ECML) method for this problem. ECML first learns view-specific evidence, which could be termed as the amount of support to each category collected from data. Then, we can construct view-specific opinions consisting of decision results and reliability. In the multi-view fusion stage, we propose a conflictive opinion aggregation strategy and theoretically prove this strategy can exactly model the relation of multi-view common and view-specific reliabilities. Experiments performed on 6 datasets verify the effectiveness of ECML. The code is released at https://github.com/jiajunsi/RCML.

show abstract

Adaptive View-Aligned and Feature Augmentation Network for Partially View-Aligned Clustering

Cited by 2 publications

References 15 publications

Continual Multi-view Clustering

Continual Multi-view Clustering

Reliable Conflictive Multi-View Learning

Contact Info

Product

Resources

About