Improving Time Series Classification with Representation Soft Label Smoothing

TL;DR

This paper introduces a novel representation soft label smoothing technique for time series classification that enhances model generalization by producing more reliable soft labels, outperforming traditional hard label training.

Contribution

The paper proposes a new representation soft label smoothing method that improves time series classification models' robustness and performance across different architectures.

Findings

Representation soft label smoothing improves classification accuracy.

The method is effective across various model structures.

Enhanced techniques outperform baseline hard label training.

Abstract

Previous research has indicated that deep neural network based models for time series classification (TSC) tasks are prone to overfitting. This issue can be mitigated by employing strategies that prevent the model from becoming overly confident in its predictions, such as label smoothing and confidence penalty. Building upon the concept of label smoothing, we propose a novel approach to generate more reliable soft labels, which we refer to as representation soft label smoothing. We apply label smoothing, confidence penalty, and our method representation soft label smoothing to several TSC models and compare their performance with baseline method which only uses hard labels for training. Our results demonstrate that the use of these enhancement techniques yields competitive results compared to the baseline method. Importantly, our method demonstrates strong performance across models with varying structures and complexities.