Deep Partial Multi-View Learning

TL;DR

This paper introduces CPM-Nets, a novel framework for multi-view learning that effectively models complex correlations among views, handles missing data, and enhances representation completeness and versatility through adversarial training and structured loss.

Contribution

The paper proposes CPM-Nets, a new multi-view learning framework that models complex view correlations, imputes missing views, and improves representation versatility with theoretical guarantees.

Findings

Outperforms existing methods in classification accuracy.

Effectively imputes missing views with stable adversarial training.

Produces structured, generalizable representations under view-missing scenarios.

Abstract

Although multi-view learning has made signifificant progress over the past few decades, it is still challenging due to the diffificulty 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 flflexibly take advantage of multiple partial views. We fifirst provide a formal defifinition 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 specififically 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 classifification loss is introduced to produce structured representations and prevent overfifitting, 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 classifification, representation learning and data imputation.