Hierarchical Optimal Transport for Robust Multi-View Learning

Abstract

Traditional multi-view learning methods often rely on two assumptions: ($i$) the samples in different views are well-aligned, and ($ii$) their representations in latent space obey the same distribution. Unfortunately, these two assumptions may be questionable in practice, which limits the application of multi-view learning. In this work, we propose a hierarchical optimal transport (HOT) method to mitigate the dependency on these two assumptions. Given unaligned multi-view data, the HOT method penalizes the sliced Wasserstein distance between the distributions of different views. These sliced Wasserstein distances are used as the ground distance to calculate the entropic optimal transport across different views, which explicitly indicates the clustering structure of the views. The HOT method is applicable to both unsupervised and semi-supervised learning, and experimental results show that it performs robustly on both synthetic and real-world tasks.

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• Multi-View Learning in the Presence of View Disagreement
• Cross Language Text Classification via Subspace Co-Regularized Multi-View Learning
• Canonical Correlation Analysis (CCA) Based Multi-View Learning: An Overview
• Distilled Wasserstein Learning for Word Embedding and Topic Modeling
• Hierarchical Optimal Transport for Multimodal Distribution Alignment
• Learning Autoencoders with Relational Regularization
• Gromov-Wasserstein Factorization Models for Graph Clustering

Cited by in corpus (2)

• On Transportation of Mini-batches: A Hierarchical Approach
• Learning Graphons via Structured Gromov-Wasserstein Barycenters