arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1607.03050cs.LGstat.ML
收藏

学习类条件KNN的度量

Learning a metric for class-conditional KNN

Janelia Research Campus, HHMI(珍利亚研究园区,霍华德·休斯医学研究所);University of Guelph(圭尔夫大学)

Daniel Jiwoong Im, Graham W. Taylor

摘要:

朴素贝叶斯最近邻(NBNN)是一个简单而有效的框架,它解决了K-最近邻(KNN)分类的许多陷阱。它在几个计算机视觉基准测试中取得了有竞争力的结果。它的核心原则是在NN搜索过程中,查询不会与数据库中的每个示例进行比较,忽略了类信息。相反,NN搜索在每个类内执行,从而为每个类生成分数。NN技术(包括NBNN)的一个关键问题是,当数据表示不能捕获感知(例如,基于类的)相似性。NBNN通过使用独立的工程描述符(例如~ SIFT)。为了将其适用性扩展到基于图像的领域之外,我们建议学习一种捕获感知相似性的度量。类似于邻域成分分析如何优化KNN分类的可微形式,我们提出了“类条件”度量学习(CCML),它优化了NBNN选择规则的软形式。典型的度量学习算法学习全局或局部度量。然而,我们提出的方法可以通过调整单个参数来调整到特定的局部水平。对分类和检索任务的实证评估表明,我们提出的方法在各种图像和非图像数据集上明显优于现有的学习距离度量。

英文摘要:

Naive Bayes Nearest Neighbour (NBNN) is a simple and effective framework which addresses many of the pitfalls of K-Nearest Neighbour (KNN) classification. It has yielded competitive results on several computer vision benchmarks. Its central tenet is that during NN search, a query is not compared to every example in a database, ignoring class information. Instead, NN searches are performed within each class, generating a score per class. A key problem with NN techniques, including NBNN, is that they fail when the data representation does not capture perceptual (e.g.~class-based) similarity. NBNN circumvents this by using independent engineered descriptors (e.g.~SIFT). To extend its applicability outside of image-based domains, we propose to learn a metric which captures perceptual similarity. Similar to how Neighbourhood Components Analysis optimizes a differentiable form of KNN classification, we propose "Class Conditional" metric learning (CCML), which optimizes a soft form of the NBNN selection rule. Typical metric learning algorithms learn either a global or local metric. However, our proposed method can be adjusted to a particular level of locality by tuning a single parameter. An empirical evaluation on classification and retrieval tasks demonstrates that our proposed method clearly outperforms existing learned distance metrics across a variety of image and non-image datasets.

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。

↑