Abstract:The state-of-the-art multi-task multi-view learning (MTMV) tackles the learning scenario where multiple tasks are associated with each other via multiple shared feature views. However, in online practical scenarios where the learning tasks have heterogeneous features collected from multiple views, e.g., multiple sources, the state-of-the-arts with single view cannot work well. To tackle this issue, in this paper, we propose a Robust Lifelong Multi-task Multi-view Representation Learning (rLM 2 L) model to accumulate the knowledge from online multi-view tasks. More specifically, we firstly design a set of view-specific libraries to maintain the intra-view correlation information of each view, and further impose an orthogonal promoting term to enforce libraries to be as independent as possible. When online new multi-view task is coming, rLM 2 L model decomposes all views of the new task into a common view-invariant space by transferring the knowledge of corresponding library. In this view-invariant space, capturing underlying inter-view correlation and identifying task-specific views for the new task are jointly employed via a robust multi-task learning formulation. Then the view-specific libraries can be refined over time to keep on improving across all tasks. For the model optimization, the proximal alternating linearized minimization algorithm is adopted to optimize our nonconvex model alternatively to achieve lifelong learning. Finally, extensive experiments on benchmark datasets shows that our proposed rLM 2 L model outperforms existing lifelong learning models, while it can discover task-specific views from sequential multi-view task with less computational burden.

Shared Structure Learning for Multiple Tasks with Multiple Views.

Multi-View Clustering Via Simultaneously Learning Shared Subspace And Affinity Matrix

Multi-View Correlated Feature Learning by Uncovering Shared Component.

Multi-task Multi-view Learning for Heterogeneous Tasks.

DMTMV: A Unified Learning Framework for Deep Multi-task Multi-view Learning.

Adaptively Sharing Multi-Levels of Distributed Representations in Multi-Task Learning

Distributed Learning of Predictive Structures from Multiple Tasks over Networks

Collaborative Structure and Feature Learning for Multi-View Clustering

Informative Relationship Multi-Task Learning: Exploring Pairwise Contribution Across Tasks' Sharing Knowledge

Learning Sparse Sharing Architectures for Multiple Tasks.

Continual Multiview Task Learning Via Deep Matrix Factorization.

TSMV: Task-Specific Multi-View Feature Learning.

Retargeted Multi-View Feature Learning with Separate and Shared Subspace Uncovering

Exploring Shared Structures and Hierarchies for Multiple NLP Tasks

Multi-View Learning with Incomplete Views

Robust Lifelong Multi-task Multi-view Representation Learning

Multi-View Matrix Decomposition: A New Scheme for Exploring Discriminative Information.

Learning Task Relational Structure for Multi-task Feature Learning

Interest Inference Via Structure-Constrained Multi-Source Multi-Task Learning

A Survey on Multi-view Learning

Efficient Computation Sharing for Multi-Task Visual Scene Understanding