Partial Adversarial Domain Adaptation

2018/08/10 by Zhangjie Cao, Lijia Ma, Cao, Zhangjie +5 · 12 citations
Computer Science · Medicine · #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Viral Infections and Vectors

paper · pdf · doi:10.48550/arxiv.1808.04205

openalex publication_date 2018/08/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Domain adversarial learning aligns the feature distributions across the source and target domains in a two-player minimax game. Existing domain adversarial networks generally assume identical label space across different domains. In the presence of big data, there is strong motivation of transferring deep models from existing big domains to unknown small domains. This paper introduces partial domain adaptation as a new domain adaptation scenario, which relaxes the fully shared label space assumption to that the source label space subsumes the target label space. Previous methods typically match the whole source domain to the target domain, which are vulnerable to negative transfer for the partial domain adaptation problem due to the large mismatch between label spaces. We present Partial Adversarial Domain Adaptation (PADA), which simultaneously alleviates negative transfer by down-weighing the data of outlier source classes for training both source classifier and domain adversary, and promotes positive transfer by matching the feature distributions in the shared label space. Experiments show that PADA exceeds state-of-the-art results for partial domain adaptation tasks on several datasets.

Citations

• Deep Domain Confusion: Maximizing for Domain Invariance
• Domain Adaptation: Learning Bounds and Algorithms
• Adversarial Discriminative Domain Adaptation

Cited by

• Bridging simulation and reality in subsurface radar-based sensing: physics-guided hierarchical domain adaptation with deep adversarial learning
• Enhancing Knowledge Transfer in Hyperspectral Image Classification via Cross-scene Knowledge Integration
• Characterizing and Avoiding Negative Transfer
• Cross-Domain Adaptation for Animal Pose Estimation
• Mitigating Negative Transfer via Reducing Environmental Disagreement
• Training-Free Label Space Alignment for Universal Domain Adaptation
• Discriminative Clustering for Robust Unsupervised Domain Adaptation
• Domain Adaptation and Image Classification via Deep Conditional Adaptation Network
• A Sample Selection Approach for Universal Domain Adaptation
• Partial Domain Adaptation via Importance Sampling-based Shift Correction
• Relation-Aware Slicing in Cross-Domain Alignment
• Source-Free Domain Adaptation via Multi-view Contrastive Learning

Related

• From Big to Small: Adaptive Learning to Partial-Set Domains
• Class Conditional Alignment for Partial Domain Adaptation
• Partial Transfer Learning with Selective Adversarial Networks
• Importance Weighted Adversarial Nets for Partial Domain Adaptation
• Learning to Transfer Examples for Partial Domain Adaptation