Towards Generating Real-World Time Series Data

TL;DR

This paper introduces RTSGAN, a novel generative framework designed to produce realistic, variable-length, and incomplete real-world time series data, addressing challenges like missing data and sequence variability.

Contribution

RTSGAN is the first framework to effectively generate real-world time series with missing data and variable lengths, combining encoding, decoding, and specialized handling of missing patterns.

Findings

RTSGAN outperforms previous methods in generating useful synthetic data.

The framework effectively models temporal dynamics and missing data patterns.

Experiments demonstrate improved downstream task performance with generated data.

Abstract

Time series data generation has drawn increasing attention in recent years. Several generative adversarial network (GAN) based methods have been proposed to tackle the problem usually with the assumption that the targeted time series data are well-formatted and complete. However, real-world time series (RTS) data are far away from this utopia, e.g., long sequences with variable lengths and informative missing data raise intractable challenges for designing powerful generation algorithms. In this paper, we propose a novel generative framework for RTS data - RTSGAN to tackle the aforementioned challenges. RTSGAN first learns an encoder-decoder module which provides a mapping between a time series instance and a fixed-dimension latent vector and then learns a generation module to generate vectors in the same latent space. By combining the generator and the decoder, RTSGAN is able to generate RTS which respect the original feature distributions and the temporal dynamics. To generate time series with missing values, we further equip RTSGAN with an observation embedding layer and a decide-and-generate decoder to better utilize the informative missing patterns. Experiments on the four RTS datasets show that the proposed framework outperforms the previous generation methods in terms of synthetic data utility for downstream classification and prediction tasks.