Concept-Based Unsupervised Domain Adaptation

2025-05-08

AI Summary

To address the degradation of interpretability and performance in Concept Bottleneck Models (CBMs) under unsupervised domain adaptation—caused by distribution shift—this paper proposes the first interpretable domain adaptation framework tailored for CBMs. Methodologically: (1) a relaxed thresholding mechanism is introduced to alleviate overly stringent concept distribution constraints; (2) an unsupervised, concept-level adversarial alignment module is designed, enabling concept alignment without target-domain concept annotations—the first such approach; and (3) a theoretically grounded joint optimization framework unifies concept learning and domain adaptation. Extensive experiments on multiple real-world datasets demonstrate that our method significantly outperforms existing CBM and domain adaptation baselines. Notably, it establishes the first benchmark performance for interpretable domain adaptation, validating both efficacy and theoretical soundness while preserving human-interpretable concept representations across domains.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Concept Bottleneck Models (CBMs) enhance interpretability by explaining predictions through human-understandable concepts but typically assume that training and test data share the same distribution. This assumption often fails under domain shifts, leading to degraded performance and poor generalization. To address these limitations and improve the robustness of CBMs, we propose the Concept-based Unsupervised Domain Adaptation (CUDA) framework. CUDA is designed to: (1) align concept representations across domains using adversarial training, (2) introduce a relaxation threshold to allow minor domain-specific differences in concept distributions, thereby preventing performance drop due to over-constraints of these distributions, (3) infer concepts directly in the target domain without requiring labeled concept data, enabling CBMs to adapt to diverse domains, and (4) integrate concept learning into conventional domain adaptation (DA) with theoretical guarantees, improving interpretability and establishing new benchmarks for DA. Experiments demonstrate that our approach significantly outperforms the state-of-the-art CBM and DA methods on real-world datasets.

Problem

• Enhance interpretability of CBMs under domain shifts
• Align concept representations across domains adaptively
• Enable unsupervised concept adaptation in target domains

Innovation

• Align concept representations via adversarial training
• Introduce relaxation threshold for domain-specific differences
• Infer concepts in target domain without labeled data