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 1905.07299cs.LGstat.ML
收藏

用于数据集复杂性评估的谱度量

Spectral Metric for Dataset Complexity Assessment

Université de Sherbrooke(舍布鲁克大学);Miovision Inc.(Miovision公司)

Frederic Branchaud-Charron, Andrew Achkar, Pierre-Marc Jodoin

摘要:

在本文中,我们提出了一个新的措施来衡量复杂的图像分类问题。给定一个带注释的图像数据集,我们的方法计算一个称为累积谱梯度(CSG)的复杂性度量,它与卷积神经网络(CNN)的测试精度密切相关。CSG度量是从谱聚类框架中类之间的概率分歧中得出的。我们表明,这个指标与数据集的整体可分性,从而其固有的复杂性。正如将要展示的,我们的度量可以用于数据集缩减,以评估哪些类更难解开,并近似于人们可以期望用CNN获得的准确度。在11个数据集和三个CNN模型上获得的结果表明,我们的方法比以前的复杂性度量更准确,更快。

英文摘要:

In this paper, we propose a new measure to gauge the complexity of image classification problems. Given an annotated image dataset, our method computes a complexity measure called the cumulative spectral gradient (CSG) which strongly correlates with the test accuracy of convolutional neural networks (CNN). The CSG measure is derived from the probabilistic divergence between classes in a spectral clustering framework. We show that this metric correlates with the overall separability of the dataset and thus its inherent complexity. As will be shown, our metric can be used for dataset reduction, to assess which classes are more difficult to disentangle, and approximate the accuracy one could expect to get with a CNN. Results obtained on 11 datasets and three CNN models reveal that our method is more accurate and faster than previous complexity measures.

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。

↑