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 2403.01966cs.CV
收藏

利用距离感知对比学习增强信息最大化,用于无源跨域小样本学习

Enhancing Information Maximization with Distance-Aware Contrastive Learning for Source-Free Cross-Domain Few-Shot Learning

University of Oulu(奥卢大学);National University of Defense Technology(国防科技大学);College of Electronic Science and Technology, National University of Defense Technology(国防科技大学电子科学学院);College of Computer, National University of Defense Technology(国防科技大学计算机学院);College of Computer Science, Nankai University(南开大学计算机学院)

Huali Xu, Li Liu, Shuaifeng Zhi, Shaojing Fu, Zhuo Su, Ming-Ming Cheng, Yongxiang Liu

AI总结:

针对无源跨域小样本学习,本文提出 IM-DCL,通过直推式信息最大化与距离感知对比软约束适配目标域,在 BSCD-FSL 的 4 个数据集上优于现有方法。

摘要:

现有的跨域小样本学习(CDFSL)方法需要在预训练阶段访问源域数据来训练模型。然而,由于人们对数据隐私的关注日益增加,并且希望降低数据传输和训练成本,因此有必要开发一种无需访问源数据的 CDFSL 解决方案。基于此,本文探讨了一个无源 CDFSL(SF-CDFSL)问题,即通过使用现有的预训练模型而不是使用源数据训练模型来解决 CDFSL,从而避免访问源数据。本文提出了一种结合距离感知对比学习的增强信息最大化(IM-DCL)方法来应对这些挑战。首先,我们引入直推式机制来学习查询集。其次,探索信息最大化(IM),将目标样本映射为个体确定性预测和全局多样性预测,帮助源模型更好地拟合目标数据分布。然而,IM 无法学习目标任务的决策边界。这促使我们引入一种称为距离感知对比学习(DCL)的新方法,其中我们将整个特征集同时视为正样本集和负样本集,类似于薛定谔的双重状态概念。我们并非在正样本集和负样本集之间进行严格分离,而是采用特征之间的加权距离计算,为整个特征集建立正样本集和负样本集的软分类。此外,我们通过在对象特征及其对应的正样本集和负样本集之间加入对比约束,解决了与 IM 相关的问题。对 BSCD-FSL 基准中 4 个数据集的评估表明,所提出的 IM-DCL 在不访问源域的情况下,表现出优于现有方法的性能,尤其是在远域任务中。

英文摘要:

Existing Cross-Domain Few-Shot Learning (CDFSL) methods require access to source domain data to train a model in the pre-training phase. However, due to increasing concerns about data privacy and the desire to reduce data transmission and training costs, it is necessary to develop a CDFSL solution without accessing source data. For this reason, this paper explores a Source-Free CDFSL (SF-CDFSL) problem, in which CDFSL is addressed through the use of existing pretrained models instead of training a model with source data, avoiding accessing source data. This paper proposes an Enhanced Information Maximization with Distance-Aware Contrastive Learning (IM-DCL) method to address these challenges. Firstly, we introduce the transductive mechanism for learning the query set. Secondly, information maximization (IM) is explored to map target samples into both individual certainty and global diversity predictions, helping the source model better fit the target data distribution. However, IM fails to learn the decision boundary of the target task. This motivates us to introduce a novel approach called Distance-Aware Contrastive Learning (DCL), in which we consider the entire feature set as both positive and negative sets, akin to Schrodinger's concept of a dual state. Instead of a rigid separation between positive and negative sets, we employ a weighted distance calculation among features to establish a soft classification of the positive and negative sets for the entire feature set. Furthermore, we address issues related to IM by incorporating contrastive constraints between object features and their corresponding positive and negative sets. Evaluations of the 4 datasets in the BSCD-FSL benchmark indicate that the proposed IM-DCL, without accessing the source domain, demonstrates superiority over existing methods, especially in the distant domain task.

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。

↑