论文标题
对齐 - 均匀性与密集对比表示的性能之间的相关性
Correlation between Alignment-Uniformity and Performance of Dense Contrastive Representations
论文作者
论文摘要
最近，与实例级对比度学习相比，密集的对比度学习在密集的预测任务上表现出了出色的表现。尽管具有至高无上的性能，但尚未仔细研究密集的对比度表示的特性。因此，我们使用标准的CNN和直接的功能匹配方案分析了密集对比度学习的理论思想，而不是提出一种新的复杂方法。灵感来自通过对超晶体的比对和均匀性的镜头对实例级对比表示的性质的分析，我们采用并扩展了相同的晶状体来分析其未流向的属性。我们发现构建积极的密集特征并经验证明其有效性的核心原则。此外，我们引入了一个新的标量指标，该指标总结了对齐和均匀性与下游性能之间的相关性。使用此指标，我们研究了密集学习的对比度表示的各个方面，例如单一和多对象数据集的相关性如何变化或线性评估以及密集的预测任务。源代码可公开可用：https：//github.com/supersupermoon/densecl-analysis
Recently, dense contrastive learning has shown superior performance on dense prediction tasks compared to instance-level contrastive learning. Despite its supremacy, the properties of dense contrastive representations have not yet been carefully studied. Therefore, we analyze the theoretical ideas of dense contrastive learning using a standard CNN and straightforward feature matching scheme rather than propose a new complex method. Inspired by the analysis of the properties of instance-level contrastive representations through the lens of alignment and uniformity on the hypersphere, we employ and extend the same lens for the dense contrastive representations to analyze their underexplored properties. We discover the core principle in constructing a positive pair of dense features and empirically proved its validity. Also, we introduces a new scalar metric that summarizes the correlation between alignment-and-uniformity and downstream performance. Using this metric, we study various facets of densely learned contrastive representations such as how the correlation changes over single- and multi-object datasets or linear evaluation and dense prediction tasks. The source code is publicly available at: https://github.com/SuperSupermoon/DenseCL-analysis
