Event Detection in Time Series: Universal Deep Learning Approach

TL;DR

This paper introduces a universal deep learning regression approach for event detection in time series, effectively handling rare events and imbalanced datasets better than traditional classification methods.

Contribution

A novel supervised regression-based deep learning method that unifies detection of various event types with fewer parameters and improved accuracy.

Findings

Outperforms traditional classification methods in diverse domains.

Effectively detects rare and imbalanced events.

Theoretically justified as a universal and precise approach.

Abstract

Event detection in time series is a challenging task due to the prevalence of imbalanced datasets, rare events, and time interval-defined events. Traditional supervised deep learning methods primarily employ binary classification, where each time step is assigned a binary label indicating the presence or absence of an event. However, these methods struggle to handle these specific scenarios effectively. To address these limitations, we propose a novel supervised regression-based deep learning approach that offers several advantages over classification-based methods. Our approach, with a limited number of parameters, can effectively handle various types of events within a unified framework, including rare events and imbalanced datasets. We provide theoretical justifications for its universality and precision and demonstrate its superior performance across diverse domains, particularly for rare events and imbalanced datasets.