Timezone: »

 
Poster
Confident Anchor-Induced Multi-Source Free Domain Adaptation
Jiahua Dong · Zhen Fang · Anjin Liu · Gan Sun · Tongliang Liu

Tue Dec 07 08:30 AM -- 10:00 AM (PST) @ Virtual
in Poster Session 1 »

Unsupervised domain adaptation has attracted appealing academic attentions by transferring knowledge from labeled source domain to unlabeled target domain. However, most existing methods assume the source data are drawn from a single domain, which cannot be successfully applied to explore complementarily transferable knowledge from multiple source domains with large distribution discrepancies. Moreover, they require access to source data during training, which are inefficient and unpractical due to privacy preservation and memory storage. To address these challenges, we develop a novel Confident-Anchor-induced multi-source-free Domain Adaptation (CAiDA) model, which is a pioneer exploration of knowledge adaptation from multiple source domains to the unlabeled target domain without any source data, but with only pre-trained source models. Specifically, a source-specific transferable perception module is proposed to automatically quantify the contributions of the complementary knowledge transferred from multi-source domains to the target domain. To generate pseudo labels for the target domain without access to the source data, we develop a confident-anchor-induced pseudo label generator by constructing a confident anchor group and assigning each unconfident target sample with a semantic-nearest confident anchor. Furthermore, a class-relationship-aware consistency loss is proposed to preserve consistent inter-class relationships by aligning soft confusion matrices across domains. Theoretical analysis answers why multi-source domains are better than a single source domain, and establishes a novel learning bound to show the effectiveness of exploiting multi-source domains. Experiments on several representative datasets illustrate the superiority of our proposed CAiDA model. The code is available at https://github.com/Learning-group123/CAiDA.

Keywords: Reinforcement Learning and Planning · Self-Supervised Learning · Domain Adaptation · Privacy · Transfer Learning

Author Information

Jiahua Dong (Shenyang Institute of Automation, Chinese Academy of Sciences)
Zhen Fang (University of Technology Sydney)
Anjin Liu (University of Technology Sydney)
Gan Sun (Chinese Academy of Sciences)
Tongliang Liu (The University of Sydney)

More from the Same Authors