Enhancing Transformer Efficiency for Multivariate Time Series Classification
This work addresses efficiency issues for practitioners handling large-scale multivariate time series data, though it appears incremental as it builds on existing transformer architectures.
The paper tackled the problem of computational inefficiency in multivariate time series classification for large-scale datasets by proposing a module-wise pruning and Pareto analysis method, resulting in reduced training time and memory footprint while maintaining accuracy.
Most current multivariate time series (MTS) classification algorithms focus on improving the predictive accuracy. However, for large-scale (either high-dimensional or long-sequential) time series (TS) datasets, there is an additional consideration: to design an efficient network architecture to reduce computational costs such as training time and memory footprint. In this work we propose a methodology based on module-wise pruning and Pareto analysis to investigate the relationship between model efficiency and accuracy, as well as its complexity. Comprehensive experiments on benchmark MTS datasets illustrate the effectiveness of our method.