Target Consistency for Domain Adaptation: when Robustness meets Transferability
This addresses domain adaptation challenges for machine learning practitioners by improving transferability, though it is incremental as it builds on existing invariant representation methods.
The paper tackled the problem that invariant representations in unsupervised domain adaptation may not ensure good target classifier performance due to violations of the cluster assumption in the target domain, and by enforcing target consistency with class-level invariance, it achieved significant improvements over state-of-the-art methods on image classification and segmentation benchmarks.
Learning Invariant Representations has been successfully applied for reconciling a source and a target domain for Unsupervised Domain Adaptation. By investigating the robustness of such methods under the prism of the cluster assumption, we bring new evidence that invariance with a low source risk does not guarantee a well-performing target classifier. More precisely, we show that the cluster assumption is violated in the target domain despite being maintained in the source domain, indicating a lack of robustness of the target classifier. To address this problem, we demonstrate the importance of enforcing the cluster assumption in the target domain, named Target Consistency (TC), especially when paired with Class-Level InVariance (CLIV). Our new approach results in a significant improvement, on both image classification and segmentation benchmarks, over state-of-the-art methods based on invariant representations. Importantly, our method is flexible and easy to implement, making it a complementary technique to existing approaches for improving transferability of representations.