Propensity-driven Uncertainty Learning for Sample Exploration in Source-Free Active Domain Adaptation

2025-01-23

AI Summary

Addressing source-free active domain adaptation (SFADA), where source-domain data is unavailable, existing methods suffer from suboptimal target-sample selection, limited noise robustness, inadequate correction of pseudo-label bias, and reliance on multiple annotation rounds. This paper proposes a propensity-driven uncertainty learning framework. It introduces a novel joint assessment mechanism—homogeneity propensity estimation and correlation indexing—to select high-quality candidate samples. A center-correlation loss is designed to jointly optimize pseudo-label fidelity and intra-class compactness. Furthermore, the framework integrates feature-level homogeneity modeling, correlation indexing, pseudo-label refinement, and self-supervised consistency regularization, enabling efficient single-round annotation and stable convergence. Extensive experiments on four benchmark datasets demonstrate significant improvements over state-of-the-art methods, achieving higher accuracy and superior robustness to label noise.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Source-free active domain adaptation (SFADA) addresses the challenge of adapting a pre-trained model to new domains without access to source data while minimizing the need for target domain annotations. This scenario is particularly relevant in real-world applications where data privacy, storage limitations, or labeling costs are significant concerns. Key challenges in SFADA include selecting the most informative samples from the target domain for labeling, effectively leveraging both labeled and unlabeled target data, and adapting the model without relying on source domain information. Additionally, existing methods often struggle with noisy or outlier samples and may require impractical progressive labeling during training. To effectively select more informative samples without frequently requesting human annotations, we propose the Propensity-driven Uncertainty Learning (ProULearn) framework. ProULearn utilizes a novel homogeneity propensity estimation mechanism combined with correlation index calculation to evaluate feature-level relationships. This approach enables the identification of representative and challenging samples while avoiding noisy outliers. Additionally, we develop a central correlation loss to refine pseudo-labels and create compact class distributions during adaptation. In this way, ProULearn effectively bridges the domain gap and maximizes adaptation performance. The principles of informative sample selection underlying ProULearn have broad implications beyond SFADA, offering benefits across various deep learning tasks where identifying key data points or features is crucial. Extensive experiments on four benchmark datasets demonstrate that ProULearn outperforms state-of-the-art methods in domain adaptation scenarios.

Problem

• Passive Active Domain Adaptation
• Sample Selection
• Fuzzy or Abnormal Data Handling

Innovation

• ProULearn
• SFADA
• data homogeneity estimation