LGSPMLDec 20, 2023

Revisiting Deep Generalized Canonical Correlation Analysis

arXiv:2312.13455v18 citationsh-index: 6IEEE Transactions on Signal Processing
Originality Incremental advance
AI Analysis

This work addresses a domain-specific problem in multiview data analysis for researchers and practitioners, offering an incremental improvement over existing deep CCA methods.

The authors tackled the limitations of existing deep canonical correlation analysis methods, such as trivial solutions and computational complexity, by proposing a novel formulation that models private components as conditionally independent given common ones, resulting in improved efficiency and identification of common factors as demonstrated in experiments.

Canonical correlation analysis (CCA) is a classic statistical method for discovering latent co-variation that underpins two or more observed random vectors. Several extensions and variations of CCA have been proposed that have strengthened our capabilities in terms of revealing common random factors from multiview datasets. In this work, we first revisit the most recent deterministic extensions of deep CCA and highlight the strengths and limitations of these state-of-the-art methods. Some methods allow trivial solutions, while others can miss weak common factors. Others overload the problem by also seeking to reveal what is not common among the views -- i.e., the private components that are needed to fully reconstruct each view. The latter tends to overload the problem and its computational and sample complexities. Aiming to improve upon these limitations, we design a novel and efficient formulation that alleviates some of the current restrictions. The main idea is to model the private components as conditionally independent given the common ones, which enables the proposed compact formulation. In addition, we also provide a sufficient condition for identifying the common random factors. Judicious experiments with synthetic and real datasets showcase the validity of our claims and the effectiveness of the proposed approach.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes