Abstract:In recent years, a great many methods of learning from multi-view data by considering the diversity of different views have been proposed. These views may be obtained from multiple sources or different feature subsets. In trying to organize and highlight similarities and differences between the variety of multi-view learning approaches, we review a number of representative multi-view learning algorithms in different areas and classify them into three groups: 1) co-training, 2) multiple kernel learning, and 3) subspace learning. Notably, co-training style algorithms train alternately to maximize the mutual agreement on two distinct views of the data; multiple kernel learning algorithms exploit kernels that naturally correspond to different views and combine kernels either linearly or non-linearly to improve learning performance; and subspace learning algorithms aim to obtain a latent subspace shared by multiple views by assuming that the input views are generated from this latent subspace. Though there is significant variance in the approaches to integrating multiple views to improve learning performance, they mainly exploit either the consensus principle or the complementary principle to ensure the success of multi-view learning. Since accessing multiple views is the fundament of multi-view learning, with the exception of study on learning a model from multiple views, it is also valuable to study how to construct multiple views and how to evaluate these views. Overall, by exploring the consistency and complementary properties of different views, multi-view learning is rendered more effective, more promising, and has better generalization ability than single-view learning.

Multi-Instance Learning: A Survey

Multi-Instance Learning from Supervised View

Semantic Image Retrieval Based on Multiple-Instance Learning

Ensembles Of Multi-Instance Learners

Multiple-Instance Learning from Pairwise Comparison Bags

Multiple instance learning: A survey of problem characteristics and applications

Multi-instance clustering with applications to multi-instance prediction

Neural Networks for Multi-Instance Learning

Multi-Instance Multi-Label Learning with Application to Scene Classification

A Survey on Multi-view Learning

Solving the Multi-Instance Problem with Neural Networks

On the relation between multi-instance learning and semi-supervised learning

Multi-Instance Learning with Distribution Change

A Review on Multi-Label Learning Algorithms

Multi-instance learning by treating instances as non-I.I.D. samples

A review of multi-instance learning assumptions

Multiple instance learning from similarity-confidence bags

A Survey of Multi-View Representation Learning

A Survey on Multi-output Learning

A New Multiple Instance Algorithm Using Structural Information.