arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.14263cs.LGstat.ML
收藏

领域自适应的目标一致性:当鲁棒性满足可移植性时

Target Consistency for Domain Adaptation: when Robustness meets Transferability

Université Paris-Saclay(巴黎萨克雷大学);CentraleSupélec(中央理工-高等电力学院);Sidetrade

Yassine Ouali, Victor Bouvier, Myriam Tami, Céline Hudelot

摘要:

学习不变表示已成功地应用于协调无监督领域自适应的源和目标域。通过研究这些方法的鲁棒性的棱镜下的集群假设,我们带来了新的证据,不变性与低源风险并不能保证一个性能良好的目标分类。更确切地说,我们表明,集群假设违反了在目标域,尽管被保持在源域,表明缺乏鲁棒性的目标分类。为了解决这个问题,我们证明了在目标域中执行集群假设的重要性,命名为目标一致性(TC),特别是当与类级不变(CLIV)配对时。我们的新方法在图像分类和分割基准上都有了显着的改进,超过了基于不变表示的最先进方法。重要的是,我们的方法是灵活的,易于实现,使其成为现有方法的补充技术,以提高表示的可转移性。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑