Unsupervised Correlation Analysis

TL;DR

This paper introduces Unsupervised Correlation Analysis (UCA), a novel method for linking different data domains without prior correspondences, using a combination of reconstruction, cycle loss, and domain confusion, achieving near-supervised performance.

Contribution

The paper presents UCA, the first unsupervised method for domain linking that does not require prior correspondences, expanding the applicability of correlation analysis.

Findings

UCA outperforms other unsupervised baselines on CCA benchmarks.

UCA approaches supervised performance in some cases.

UCA successfully links remote domains like text and images.

Abstract

Linking between two data sources is a basic building block in numerous computer vision problems. In this paper, we set to answer a fundamental cognitive question: are prior correspondences necessary for linking between different domains? One of the most popular methods for linking between domains is Canonical Correlation Analysis (CCA). All current CCA algorithms require correspondences between the views. We introduce a new method Unsupervised Correlation Analysis (UCA), which requires no prior correspondences between the two domains. The correlation maximization term in CCA is replaced by a combination of a reconstruction term (similar to autoencoders), full cycle loss, orthogonality and multiple domain confusion terms. Due to lack of supervision, the optimization leads to multiple alternative solutions with similar scores and we therefore introduce a consensus-based mechanism that is often able to recover the desired solution. Remarkably, this suffices in order to link remote domains such as text and images. We also present results on well accepted CCA benchmarks, showing that performance far exceeds other unsupervised baselines, and approaches supervised performance in some cases.