Multivariate time series classification with dual attention network

Mojtaba A. Farahani,Tara Eslaminokandeh

2023-08-27

Abstract:One of the topics in machine learning that is becoming more and more relevant is multivariate time series classification. Current techniques concentrate on identifying the local important sequence segments or establishing the global long-range dependencies. They frequently disregard the merged data from both global and local features, though. Using dual attention, we explore a novel network (DA-Net) in this research to extract local and global features for multivariate time series classification. The two distinct layers that make up DA-Net are the Squeeze-Excitation Window Attention (SEWA) layer and the Sparse Self-Attention within Windows (SSAW) layer. DA- Net can mine essential local sequence fragments that are necessary for establishing global long-range dependencies based on the two expanded layers.

Machine Learning

What problem does this paper attempt to address?

### Problems Addressed by the Paper The paper primarily addresses the issue of Multivariate Time Series Classification (MTSC). Specifically: 1. **Limitations of Existing Techniques**: Current techniques mainly focus on identifying locally important sequence segments or establishing global long-range dependencies, but often overlook the combination of global and local features. 2. **Proposed New Method**: By introducing a Dual Attention mechanism, the authors propose a new network structure (DA-Net) to extract both local and global features in multivariate time series. DA-Net includes two distinct layers: the Squeeze-Excitation Window Attention (SEWA) layer and the Sparse Self-Attention within Windows (SSAW) layer. 3. **Addressing Long Sequence Issues**: The SEWA layer explores the relationships between windows through global window features and dynamically calibrates features to learn local characteristics; the SSAW layer reduces computational complexity within windows through a sparse attention mechanism, enabling the network to generalize and uncover global long-range dependencies. In summary, the paper aims to provide an efficient and accurate method for handling multivariate time series classification problems through DA-Net.

Multivariate time series classification with dual attention network

Multidimensional Time Series Classification with Multiple Attention Mechanism

Multisource Remote Sensing Data Classification Based on A Dual Attention Fusion Network

Attention-Aware Deep Feature Embedding for Remote Sensing Image Scene Classification

CANet: Convolutional Attention-based Network for Multivariate Time Series Anomaly Detection

DSDCLNet: Dual-stream encoder and dual-level contrastive learning network for supervised multivariate time series classification

SiamDA: Dual Attention Siamese Network for Real-Time Visual Tracking

Attention Based CNN-LSTM Network for Anomaly Pattern Classification of Multivariate Time Series

Improving Irregularly Sampled Time Series Learning with Time-Aware Dual-Attention Memory-Augmented Networks.

Self-Attention Causal Dilated Convolutional Neural Network for Multivariate Time Series Classification and Its Application

DTAAD: Dual Tcn-attention networks for anomaly detection in multivariate time series data

Exploiting Multi-Channels Deep Convolutional Neural Networks for Multivariate Time Series Classification

TapNet: Multivariate Time Series Classification with Attentional Prototypical Network

A Multivariate Time Series Prediction Method Based on Convolution-Residual Gated Recurrent Neural Network and Double-Layer Attention

Attention-Based Deep Gated Fully Convolutional End-to-End Architectures for Time Series Classification

Multistage attention network for multivariate time series prediction

Spatiotemporal Learning with Decoupled Causal Attention for Multivariate Time Series

TV-Net: Temporal-Variable feature harmonizing Network for multivariate time series classification and interpretation

End-to-end Multivariate Time Series Classification Via Hybrid Deep Learning Architectures

Multivariate Time Series Classification: A Deep Learning Approach