Deep Partial Multi-View Learning

By Changqing Zhang, Yajie Cui, Zongbo Han, Joey Tianyi Zhou, Huan Fu, Qinghua Hu. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020

315 citations recorded by the source corpus.

Although multi-view learning has made significant progress over the past few decades, it is still challenging due to the difficulty in modeling complex correlations among different views, especially under the context of view missing. To address the challenge, we propose a novel framework termed Cross Partial Multi-View Networks (CPM-Nets), which aims to fully and flexibly take advantage of multiple partial views. We first provide a formal definition of completeness and versatility for multi-view representation and then theoretically prove the versatility of the learned latent representations. For completeness, the task of learning latent multi-view representation is specifically translated to a degradation process by mimicking data transmission, such that the optimal tradeoff between consistency and complementarity across different views can be implicitly achieved. Equipped with adversarial strategy, our model stably imputes missing views, encoding information from all views for each sample to be encoded into latent representation to further enhance the completeness. Furthermore, a nonparametric classification loss is introduced to produce structured representations and prevent overfitting, which endows the algorithm with promising generalization under view-missing cases. Extensive experimental results validate the effectiveness of our algorithm over existing state of the arts for classification, representation learning and data imputation.

Subfields: Artificial Intelligence, Computational Mechanics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/35302ad4f7bffeb60bf68a292d462adac6f02780

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jiancheng Lv
2. Joey Tianyi Zhou
3. Zongbo Han
4. Peng Hu
5. Xi Peng
6. Changqing Zhang
7. Yuanbiao Gou
8. Huan Fu

In Computational Mechanics

1. Joey Tianyi Zhou
2. Xi Peng
3. Changqing Zhang
4. Yuanbiao Gou
5. Huan Fu
6. Jiancheng Lv
7. Yijie Lin

In Computer Vision and Pattern Recognition

1. Changqing Zhang
2. Huan Fu
3. Jiancheng Lv
4. Joey Tianyi Zhou
5. Zongbo Han
6. Peng Hu
7. Xi Peng
8. Yuanbiao Gou

Citations in the corpus

• Trusted Multi-View Classification With Dynamic Evidential Fusion (2022, methodology): “These methods are typically built on complex models [1], [2], [3], [4], [5], [6], [7], [8], [9] (usually integrating different views with deep neural networks (DNNs) ) and rely on large datasets with semantic annotations.”
• Robust Multi-View Clustering With Incomplete Information (2022, methodology): “During past years, some efforts have been devoted to solving PSP by imputing the missing samples with various data recovery methods [5], [6], [7].”
• Dual Contrastive Prediction for Incomplete Multi-View Representation Learning (2022, methodology): “(12) CPMNets (Cross Partial Multi-View Networks) [14] classifies partial multi-view data by focusing on the completeness and versatility of learned representations.”