Sig-Wasserstein GANs for Time Series Generation
This addresses the need for synthetic time series generation to accelerate AI/ML pipelines, particularly in finance, though it appears incremental as it builds on existing GAN and signature methods.
The paper tackles the problem of generating high-fidelity synthetic time series data by developing SigWGAN, which combines continuous-time stochastic models with a signature-based Wasserstein metric, turning the GAN min-max problem into supervised learning and achieving high-fidelity samples validated on synthetic and financial data.
Synthetic data is an emerging technology that can significantly accelerate the development and deployment of AI machine learning pipelines. In this work, we develop high-fidelity time-series generators, the SigWGAN, by combining continuous-time stochastic models with the newly proposed signature $W_1$ metric. The former are the Logsig-RNN models based on the stochastic differential equations, whereas the latter originates from the universal and principled mathematical features to characterize the measure induced by time series. SigWGAN allows turning computationally challenging GAN min-max problem into supervised learning while generating high fidelity samples. We validate the proposed model on both synthetic data generated by popular quantitative risk models and empirical financial data. Codes are available at https://github.com/SigCGANs/Sig-Wasserstein-GANs.git.