Abstract:Learning features from multiple views has attracted much research attention in different machine learning tasks, such as multiclass and multilabel classification problems. In this paper, we propose a multiclass multilabel multiview learning framework with a linear computational cost where an example is associated with at least one label and represented by multiple information sources. We simultaneously analyze all features by learning an integrated projection matrix. We can also automatically select more important views for subsequent classifier to predict each class. As the proposed objective function is nonsmooth and difficult to solve, we apply a novel optimization method that converts the multiview learning problem to a set of linear single-view learning problems by bridging our problem to an easily solvable approach. Compared to the conventional methods which learn the entire projection matrix, our algorithm independently optimizes each column of the projection matrix for each class, which can be easily parallelized. In each column optimization, the most computationally intensive step is pure and simple matrix-by-vector multiplication. As a result, our algorithm is much more applicable to large-scale problems than the multiview learning methods with a nonlinear computational cost. Moreover, rigorous convergence proof of the proposed algorithm is also provided. To evaluate the effectiveness of the proposed approach, experimental comparisons are made with state-of-the-art algorithms in multiclass and multilabel classification tasks on many multiview benchmarks. We also report the efficiency comparison results on different numbers of data samples. The experimental results demonstrate that our algorithm can achieve superior performance to all the compared algorithms.

Learning View-Specific Labels and Label-Feature Dependence Maximization for Multi-View Multi-Label Classification.

A Multiview Learning Framework with a Linear Computational Cost

Multi-View Analysis Dictionary Learning for Image Classification.

Common-Individual Semantic Fusion for Multi-View Multi-Label Learning

Deep Double Incomplete Multi-View Multi-Label Learning With Incomplete Labels and Missing Views

Incomplete Multi-View Weak-Label Learning with Noisy Features and Imbalanced Labels

Robust Mapping Learning for Multi-view Multi-label Classification with Missing Labels.

Partial Multi-View Multi-Label Classification Via Semantic Invariance Learning and Prototype Modeling

Incomplete Multi-View Multi-Label Learning via Label-Guided Masked View- and Category-Aware Transformers

LCBM: A Multi-View Probabilistic Model for Multi-Label Classification

Global and local multi-view multi-label learning with incomplete views and labels

Multi-scale locality preserving projection for partial multi-view incomplete multi-label learning

Label-Specific Alignment with Adversarial Multi-View Representation

Embedding Regularizer Learning for Multi-View Semi-Supervised Classification.

A Reconstruction Error Based Framework for Multi-Label and Multi-View Learning

Tensor Based Multi-View Label Enhancement for Multi-Label Learning

Incomplete Multi-view Multi-label Classification via a Dual-level Contrastive Learning Framework

Individuality- and Commonality-Based Multiview Multilabel Learning

Reliable Representations Learning for Incomplete Multi-View Partial Multi-Label Classification

Multi-view multi-label learning for image annotation

Learning with Weak Views Based on Dependence Maximization Dimensionality Reduction.