Online Bayesian max-margin subspace learning for multi-view classification and regression

Abstract

Multi-view data have become increasingly popular in many real-world applications where data are generated from different information channels or different views such as image + text, audio + video, and webpage + link data. Last decades have witnessed a number of studies devoted to multi-view learning algorithms, especially the predictive latent subspace learning approaches which aim at obtaining a subspace shared by multiple views and then learning models in the shared subspace. However, few efforts have been made to handle online multiview learning scenarios. In this paper, we propose an online Bayesian multi-view learning algorithm which learns predictive subspace with the max-margin principle. Specifically, we first define the latent margin loss for classification or regression in the subspace, and then cast the learning problem into a variational Bayesian framework by exploiting the pseudolikelihood and data augmentation idea. With the variational approximate posterior inferred from the past samples, we can naturally combine historical knowledge with new arrival data, in a Bayesian passive-aggressive style. Finally, we extensively evaluate our model on several real-world data sets and the experimental results show that our models can achieve superior performance, compared with a number of state-of-the-art competitors. Keywords Multi-view learning • Online learning • Bayesian subspace learning • Max-margin • Classification • Regression Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Fusion of a Static and Dynamic Convolutional Neural Network for Multiview 3D Point Cloud Classification

“…The network model can integrate multiview and feature data in multilevel, and the calculation of relevant parameters is repeatable and in line with common sense; however, the complexity of the model is high and its calculations are inefficient. He et al [38] presented an online Bayesian multiview learning algorithm that learns the prediction subspace based on the principle of maximum margin, defines the potential marginal loss, and minimizes the learning problems associ-ated with various Bayesian frameworks by using the theory of pseudo-likelihood and data enhancement. It obtains an approximate a posteriori change according to past samples.…”

Deep Co-Attention Network for Multi-View Subspace Learning

“…In multiple kernel learning, [5] proposed a two-view Support Vector Machine method (SVM-2K) to find multiple kernels to maximize the correlation of the two views; [28] proposed a coregularization method to jointly regularize two Reproducing Kernel Hilbert Spaces associated with the two views; [1] proposed Deep Canonical Correlation Analysis to find two deep networks such that the output layers of the two networks are maximally correlated. As for the subspace learning, the authors of [14] proposed a deep multiview robust representation learning algorithm based on auto-encoder to learn a shared representation from multi-view observations; [11] proposed online Bayesian subspace multi-view learning by modeling the variational approximate posterior inferred from the past samples; [51] proposed M2VW for multi-view multi-worker learning problem by leveraging the structural information between multiple views and multiple workers; [30] proposed CR-GAN method to learn a complete representation for multi-view generations in the adversarial setting by the collaboration of two learning pathways in a parameter-sharing manner. Different from [30,[41][42][43][44][50][51][52][53], in this paper, we focus on multi-view classification problem and aim to extract both the shared information and the view-specific information in the adversarial setting, and the view consistency constraint with label information is utilized to further regularize the generated representation in order to improve the predictive performance.…”

MV-GCN: Multi-View Graph Convolutional Networks for Link Prediction

“…Some detailed surveys about this framework can be found in [29], [40]. A large portion of multi-view learning literatures mainly focus on supervised (or semi-supervised settings) learning tasks, such as recommendation [41] and classification [42], [43]. Besides, Some other works consider an unsupervised setting.…”