Noise Optimized Conditional Diffusion for Domain Adaptation

TL;DR

This paper introduces NOCDDA, a novel domain adaptation method that leverages class-aware diffusion models to improve pseudo-label quality and cross-domain alignment, significantly outperforming existing techniques.

Contribution

The paper proposes a noise-optimized conditional diffusion approach for domain adaptation, integrating generative models with classifiers and introducing class-aware noise refinement for better alignment.

Findings

Achieves superior performance on 29 DA tasks across 5 datasets.

Outperforms 31 state-of-the-art methods in domain adaptation.

Enhances cross-domain alignment through class-aware noise optimization.

Abstract

Pseudo-labeling is a cornerstone of Unsupervised Domain Adaptation (UDA), yet the scarcity of High-Confidence Pseudo-Labeled Target Domain Samples (\textbf{hcpl-tds}) often leads to inaccurate cross-domain statistical alignment, causing DA failures. To address this challenge, we propose \textbf{N}oise \textbf{O}ptimized \textbf{C}onditional \textbf{D}iffusion for \textbf{D}omain \textbf{A}daptation (\textbf{NOCDDA}), which seamlessly integrates the generative capabilities of conditional diffusion models with the decision-making requirements of DA to achieve task-coupled optimization for efficient adaptation. For robust cross-domain consistency, we modify the DA classifier to align with the conditional diffusion classifier within a unified optimization framework, enabling forward training on noise-varying cross-domain samples. Furthermore, we argue that the conventional \( \mathcal{N}(\mathbf{0}, \mathbf{I}) \) initialization in diffusion models often generates class-confused hcpl-tds, compromising discriminative DA. To resolve this, we introduce a class-aware noise optimization strategy that refines sampling regions for reverse class-specific hcpl-tds generation, effectively enhancing cross-domain alignment. Extensive experiments across 5 benchmark datasets and 29 DA tasks demonstrate significant performance gains of \textbf{NOCDDA} over 31 state-of-the-art methods, validating its robustness and effectiveness.