Multi-view clustering via adversarial view embedding and adaptive view fusion
- PDF / 1,956,280 Bytes
- 12 Pages / 595.224 x 790.955 pts Page_size
- 62 Downloads / 151 Views
Multi-view clustering via adversarial view embedding and adaptive view fusion Yongzhen Li1,2
· Husheng Liao1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Multi-view clustering, which explores complementarity and consistency among multiple distinct feature sets to boost clustering performance, is becoming more and more useful in many real-world applications. Traditional approaches usually map multiple views to a unified embedding, in which some weighted mechanisms are utilized to measure the importance of each view. The embedding, serving as a clustering friendly representation, is then sent to extra clustering algorithms. However, a unified embedding cannot cover both complementarity and consistency among views and the weighted scheme measuring the importance of each view as a whole ignores the differences of features in each view. Moreover, because of lacking in proper grouping structure constraint imposed on the unified embedding, it will lead to just multi-view representation learned, which is not clustering friendly. In this paper, we propose a novel multi-view clustering method to alleviate the above problems. By dividing the embedding of a view into unified and view-specific vectors explicitly, complementarity and consistency can be reflected. Besides, an adversarial learning process is developed to force the above embeddings to be non-trivial. Then a fusion strategy is automatically learned, which will adaptively adjust weights for all the features in each view. Finally, a Kullback-Liebler (KL) divergence based objective is developed to constrain the fused embedding for clustering friendly representation learning and to conduct clustering. Extensive experiments have been conducted on various datasets, performing better than the state-of-the-art clustering approaches. Keywords Multi-view clustering · Adversarial view embedding · Adaptive view fusion · Clustering friendly representation learning
1 Introduction In real-world applications, it is often the case that data consist of multiple feature representations, which are called multi-view data with each view representing a feature set [1]. Usually, multi-view data represent complementarity and consistency. An example is shown in Fig. 1, content of “Canton Tower” is expressed by the image view and text
Husheng Liao
[email protected] Yongzhen Li [email protected] 1
Information department, Beijing University of Technology, Beijing, China
2
School of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, Beijing, China
view, where the tower and the Pearl River consist of the image view and the name, location, designer of the tower composed of the text view. The orange rectangles covered in both image and text views show the consistency, whereas other contents representing different aspects display the complementarity. Due to above characteristics, multi-view learning is widely studied and used for various applications, e.g. cross-modal retrieval in cross-media int
Data Loading...
