TSGM: Regular and Irregular Time-series Generation using Score-based Generative Models

TL;DR

This paper introduces TSGM, a score-based generative model tailored for synthesizing both regular and irregular time-series data with high diversity and quality, leveraging a novel conditional score network and denoising loss.

Contribution

The paper develops a flexible conditional score network for time-series synthesis, capable of handling both regular and irregular data with minimal modifications.

Findings

Achieves state-of-the-art diversity and quality in time-series synthesis

Successfully synthesizes both regular and irregular time-series datasets

Demonstrates superior performance over existing methods

Abstract

Score-based generative models (SGMs) have demonstrated unparalleled sampling quality and diversity in numerous fields, such as image generation, voice synthesis, and tabular data synthesis, etc. Inspired by those outstanding results, we apply SGMs to synthesize time-series by learning its conditional score function. To this end, we present a conditional score network for time-series synthesis, deriving a denoising score matching loss tailored for our purposes. In particular, our presented denoising score matching loss is the conditional denoising score matching loss for time-series synthesis. In addition, our framework is such flexible that both regular and irregular time-series can be synthesized with minimal changes to our model design. Finally, we obtain exceptional synthesis performance on various time-series datasets, achieving state-of-the-art sampling diversity and quality.