arXiv 2506.120422025-05-24cs.LGcs.AIstat.ML

CRITS：用于可解释时间序列分类的卷积整流器

CRITS: Convolutional Rectifier for Interpretable Time Series Classification

Department of Computer and Systems Sciences, Stockholm University（斯德哥尔摩大学计算机与系统科学系）；Department of Computer Science and Engineering, University of Oviedo（奥维多大学计算机科学与工程系）

Alejandro Kuratomi, Zed Lee, Guilherme Dinis Chaliane Junior, Tony Lindgren, Diego García Pérez

备注：: This paper was presented at the 2024 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), as part of the XKDD workshop on interpretability. However it was not published in the LNCSI proceedings of the conference

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对现有卷积网络可解释方法存在输入空间解释不足或需随机扰动的问题，提出可解释时间序列分类模型CRITS，借助全连接整流网络的ReLU激活内在提取局部解释，无需梯度、扰动和显著性图升尺度，并验证了其分类性能与解释质量。

摘要：

目前已有多种针对基于卷积网络的分类器的可解释性方法。这类方法大多聚焦于为给定样本提取显著性图，提供突出分类关键区域的局部解释。然而，其中部分方法因升尺度问题在输入空间中缺乏详细解释，或可能需要通过随机扰动来提取解释。我们提出用于可解释时间序列分类的卷积整流器（Convolutional Rectifier for Interpretable Time Series Classification，简称CRITS），这是一种专为内在提取局部解释而设计的时间序列分类可解释模型。该方法使用一层卷积核、一个最大池化层以及一个全连接整流网络（仅含修正线性单元激活的网络）。修正线性单元激活可提取给定样本的特征权重，无需计算梯度、使用随机扰动，也无需将显著性图升尺度至初始输入空间。我们在一组数据集上评估了CRITS，并研究了其分类性能以及解释的对齐度、灵敏度和可理解性。

英文摘要：

Several interpretability methods for convolutional network-based classifiers exist. Most of these methods focus on extracting saliency maps for a given sample, providing a local explanation that highlights the main regions for the classification. However, some of these methods lack detailed explanations in the input space due to upscaling issues or may require random perturbations to extract the explanations. We propose Convolutional Rectifier for Interpretable Time Series Classification, or CRITS, as an interpretable model for time series classification that is designed to intrinsically extract local explanations. The proposed method uses a layer of convolutional kernels, a max-pooling layer and a fully-connected rectifier network (a network with only rectified linear unit activations). The rectified linear unit activation allows the extraction of the feature weights for the given sample, eliminating the need to calculate gradients, use random perturbations and the upscale of the saliency maps to the initial input space. We evaluate CRITS on a set of datasets, and study its classification performance and its explanation alignment, sensitivity and understandability.

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。