dCAM: Dimension-wise Class Activation Map for Explaining Multivariate Data Series Classification

Abstract

Data series classification is an important and challenging problem in data science. Explaining the classification decisions by finding the discriminant parts of the input that led the algorithm to some decisions is a real need in many applications. Convolutional neural networks perform well for the data series classification task; though, the explanations provided by this type of algorithm are poor for the specific case of multivariate data series. Addressing this important limitation is a significant challenge. In this paper, we propose a novel method that solves this problem by highlighting both the temporal and dimensional discriminant information. Our contribution is two-fold: we first describe a convolutional architecture that enables the comparison of dimensions; then, we propose a method that returns dCAM, a Dimension-wise Class Activation Map specifically designed for multivariate time series (and CNN-based models). Experiments with several synthetic and real datasets demonstrate that dCAM is not only more accurate than previous approaches, but the only viable solution for discriminant feature discovery and classification explanation in multivariate time series. This paper has appeared in SIGMOD'22.

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• Deep learning for time series classification: a review
• Empirical Evaluation of Rectified Activations in Convolutional Network
• ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernels
• Benchmarking Multivariate Time Series Classification Algorithms
• Multi-Scale Convolutional Neural Networks for Time Series Classification
• TS-CHIEF: A Scalable and Accurate Forest Algorithm for Time Series Classification
• Proximity Forest: An effective and scalable distance-based classifier for time series
• Evaluating surgical skills from kinematic data using convolutional neural networks
• Series2Graph: Graph-based Subsequence Anomaly Detection for Time Series
• Towards a universal neural network encoder for time series
• Benchmarking Deep Learning Interpretability in Time Series Predictions
• Hercules Against Data Series Similarity Search
• Return of the Lernaean Hydra: Experimental Evaluation of Data Series Approximate Similarity Search
• The Lernaean Hydra of Data Series Similarity Search: An Experimental Evaluation of the State of the Art
• Deep learning for time series classification
• Ensemble Grammar Induction For Detecting Anomalies in Time Series

Cited by in corpus (8)

• Hercules Against Data Series Similarity Search
• Appliance Detection Using Very Low-Frequency Smart Meter Time Series
• Robust Explainer Recommendation for Time Series Classification
• ADF & TransApp: A Transformer-Based Framework for Appliance Detection Using Smart Meter Consumption Series
• DeviceScope: An Interactive App to Detect and Localize Appliance Patterns in Electricity Consumption Time Series
• Few Labels are all you need: A Weakly Supervised Framework for Appliance Localization in Smart-Meter Series
• MSAD: A Deep Dive into Model Selection for Time series Anomaly Detection
• ProtoTSNet: Interpretable Multivariate Time Series Classification With Prototypical Parts