Secure Domain Adaptation with Multiple Sources
This addresses privacy and security concerns in distributed source domains for MUDA, offering a solution for scenarios with data sharing limitations.
The paper tackles the problem of multi-source unsupervised domain adaptation (MUDA) under data privacy constraints where source domain data cannot be shared, by developing an algorithm that aligns distributions indirectly via feature embeddings and confidence-based model predictions, demonstrating effectiveness through empirical experiments.
Multi-source unsupervised domain adaptation (MUDA) is a framework to address the challenge of annotated data scarcity in a target domain via transferring knowledge from multiple annotated source domains. When the source domains are distributed, data privacy and security can become significant concerns and protocols may limit data sharing, yet existing MUDA methods overlook these constraints. We develop an algorithm to address MUDA when source domain data cannot be shared with the target or across the source domains. Our method is based on aligning the distributions of source and target domains indirectly via estimating the source feature embeddings and predicting over a confidence based combination of domain specific model predictions. We provide theoretical analysis to support our approach and conduct empirical experiments to demonstrate that our algorithm is effective.