Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation

TL;DR

This paper introduces a recursive conditional Gaussian prior for unsupervised domain adaptation in ordinal classification tasks, effectively modeling ordered label constraints and improving domain alignment in medical diagnosis and age estimation.

Contribution

It proposes a novel recursively conditional Gaussian prior for ordinal UDA, explicitly modeling label order constraints and disentangling shared and domain-specific features.

Findings

Effective in medical diagnosis domain adaptation

Improves age estimation accuracy

Outperforms existing methods in experiments

Abstract

The unsupervised domain adaptation (UDA) has been widely adopted to alleviate the data scalability issue, while the existing works usually focus on classifying independently discrete labels. However, in many tasks (e.g., medical diagnosis), the labels are discrete and successively distributed. The UDA for ordinal classification requires inducing non-trivial ordinal distribution prior to the latent space. Target for this, the partially ordered set (poset) is defined for constraining the latent vector. Instead of the typically i.i.d. Gaussian latent prior, in this work, a recursively conditional Gaussian (RCG) set is adapted for ordered constraint modeling, which admits a tractable joint distribution prior. Furthermore, we are able to control the density of content vector that violates the poset constraints by a simple "three-sigma rule". We explicitly disentangle the cross-domain images into a shared ordinal prior induced ordinal content space and two separate source/target ordinal-unrelated spaces, and the self-training is worked on the shared space exclusively for ordinal-aware domain alignment. Extensive experiments on UDA medical diagnoses and facial age estimation demonstrate its effectiveness.