Domain Adaptation for Multi-label Image Classification: a Discriminator-free Approach

2025-05-20

AI Summary

This paper addresses unsupervised domain adaptation (UDA) for multi-label image classification. We propose DDA-MLIC, an adversarial learning framework that eliminates the need for auxiliary discriminators. Our method introduces three key innovations: (1) an endogenous adversarial evaluation mechanism grounded in task-specific classifier architecture—replacing conventional external discriminators; (2) a deep neural network-based, end-to-end parameterized two-component Gaussian mixture model (GMM), circumventing iterative EM estimation; and (3) a differentiable, lightweight Fréchet distance formulation for the adversarial loss. Evaluated on three multi-label benchmarks exhibiting distinct domain shifts, DDA-MLIC achieves state-of-the-art accuracy while significantly reducing model parameters. The source code is publicly available.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

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: Humans versus LLMs for data annotation and labeling

Abstract

This paper introduces a discriminator-free adversarial-based approach termed DDA-MLIC for Unsupervised Domain Adaptation (UDA) in the context of Multi-Label Image Classification (MLIC). While recent efforts have explored adversarial-based UDA methods for MLIC, they typically include an additional discriminator subnet. Nevertheless, decoupling the classification and the discrimination tasks may harm their task-specific discriminative power. Herein, we address this challenge by presenting a novel adversarial critic directly derived from the task-specific classifier. Specifically, we employ a two-component Gaussian Mixture Model (GMM) to model both source and target predictions, distinguishing between two distinct clusters. Instead of using the traditional Expectation Maximization (EM) algorithm, our approach utilizes a Deep Neural Network (DNN) to estimate the parameters of each GMM component. Subsequently, the source and target GMM parameters are leveraged to formulate an adversarial loss using the Fr'echet distance. The proposed framework is therefore not only fully differentiable but is also cost-effective as it avoids the expensive iterative process usually induced by the standard EM method. The proposed method is evaluated on several multi-label image datasets covering three different types of domain shift. The obtained results demonstrate that DDA-MLIC outperforms existing state-of-the-art methods in terms of precision while requiring a lower number of parameters. The code is made publicly available at github.com/cvi2snt/DDA-MLIC.

Problem

• Develop discriminator-free UDA for multi-label image classification
• Replace EM with DNN for efficient GMM parameter estimation
• Improve precision with fewer parameters in domain adaptation

Innovation

• Discriminator-free adversarial approach for domain adaptation
• Gaussian Mixture Model with DNN parameter estimation
• Adversarial loss using Fréchet distance for efficiency