Dynamic Data Augmentation with Gating Networks for Time Series Recognition

TL;DR

This paper introduces a neural network with a gating mechanism that dynamically selects optimal data augmentation strategies for time series recognition, improving generalization across diverse datasets.

Contribution

It presents a novel gating network that adaptively combines data augmentation methods using feature consistency loss, tailored for time series classification.

Findings

Improves accuracy on 12 UCR time series datasets

Reveals relationships between augmentation methods

Demonstrates adaptive augmentation selection effectiveness

Abstract

Data augmentation is a technique to improve the generalization ability of machine learning methods by increasing the size of the dataset. However, since every augmentation method is not equally effective for every dataset, you need to select an appropriate method carefully. We propose a neural network that dynamically selects the best combination of data augmentation methods using a mutually beneficial gating network and a feature consistency loss. The gating network is able to control how much of each data augmentation is used for the representation within the network. The feature consistency loss gives a constraint that augmented features from the same input should be in similar. In experiments, we demonstrate the effectiveness of the proposed method on the 12 largest time-series datasets from 2018 UCR Time Series Archive and reveal the relationships between the data augmentation methods through analysis of the proposed method.