arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2501.01142cs.CV
收藏

面向多源域自适应的自适应硬度驱动增强与对齐策略

Adaptive Hardness-driven Augmentation and Alignment Strategies for Multi-Source Domain Adaptations

School of Computer Science, Sichuan University(四川大学计算机学院)

Yang Yuxiang, Zeng Xinyi, Zeng Pinxian, Zu Chen, Yan Binyu, Zhou Jiliu, Wang Yan

AI总结:

本文提出多源域自适应方法A3MDA,通过三种自适应硬度度量同时改进数据增强、域间簇级对齐和域内伪对比对齐,在多个基准上取得更优性能。

摘要:

多源域自适应(MDA)旨在将知识从多个带标签的源域迁移到无标签的目标域。然而,传统方法主要侧重于通过样本级约束(如最大均值差异 MMD)实现域间对齐,忽略了三个关键方面:1)数据增强的潜力,2)域内对齐的重要性,以及3)簇级约束的设计。本文针对 MDA 任务提出了一种新颖的硬度驱动策略,命名为“A3MDA”,通过在数据增强和域对齐中自适应地量化和利用硬度,综合考量上述三个方面。为此,“A3MDA”逐步提出了三种自适应硬度度量(AHM),即基础 AHM、平滑 AHM 和比较 AHM,每种度量都包含适用于不同场景的不同机制。具体而言,基础 AHM 旨在衡量每个源域/目标域样本的瞬时硬度。然后,由平滑 AHM 测得的硬度值将自适应地调整强数据增强的强度水平,以保持与模型泛化能力的兼容性。相比之下,比较 AHM 被设计用于促进簇级约束。通过利用硬度值作为样本特定的权重,传统的 MMD 被增强为加权聚类变体,从而增强了域间对齐的鲁棒性和精确性。对于常被忽视的域内对齐,我们通过基于硬度排名选择更难的样本来自适应地构建伪对比矩阵,提升伪标签的质量,并塑造一个聚类良好的目标特征空间。在多个 MDA 基准上的实验表明,“A3MDA”优于其他方法。

英文摘要:

Multi-source Domain Adaptation (MDA) aims to transfer knowledge from multiple labeled source domains to an unlabeled target domain. Nevertheless, traditional methods primarily focus on achieving inter-domain alignment through sample-level constraints, such as Maximum Mean Discrepancy (MMD), neglecting three pivotal aspects: 1) the potential of data augmentation, 2) the significance of intra-domain alignment, and 3) the design of cluster-level constraints. In this paper, we introduce a novel hardness-driven strategy for MDA tasks, named "A3MDA" , which collectively considers these three aspects through Adaptive hardness quantification and utilization in both data Augmentation and domain Alignment.To achieve this, "A3MDA" progressively proposes three Adaptive Hardness Measurements (AHM), i.e., Basic, Smooth, and Comparative AHMs, each incorporating distinct mechanisms for diverse scenarios. Specifically, Basic AHM aims to gauge the instantaneous hardness for each source/target sample. Then, hardness values measured by Smooth AHM will adaptively adjust the intensity level of strong data augmentation to maintain compatibility with the model's generalization capacity.In contrast, Comparative AHM is designed to facilitate cluster-level constraints. By leveraging hardness values as sample-specific weights, the traditional MMD is enhanced into a weighted-clustered variant, strengthening the robustness and precision of inter-domain alignment. As for the often-neglected intra-domain alignment, we adaptively construct a pseudo-contrastive matrix by selecting harder samples based on the hardness rankings, enhancing the quality of pseudo-labels, and shaping a well-clustered target feature space. Experiments on multiple MDA benchmarks show that " A3MDA " outperforms other methods.

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。

↑