Time Series Synthesis via Multi-scale Patch-based Generation of Wavelet Scalogram
This addresses time series synthesis for domains with limited data, but it is incremental as it adapts existing single-image generative models to wavelet scalograms.
The paper tackles unconditional generation of synthetic time series from a single sample in low-data regimes by learning patch distributions in wavelet scalograms, showing effectiveness in fidelity and diversity for time series with minimal trends, with better performance for same-duration generation than longer ones.
A framework is proposed for the unconditional generation of synthetic time series based on learning from a single sample in low-data regime case. The framework aims at capturing the distribution of patches in wavelet scalogram of time series using single image generative models and producing realistic wavelet coefficients for the generation of synthetic time series. It is demonstrated that the framework is effective with respect to fidelity and diversity for time series with insignificant to no trends. Also, the performance is more promising for generating samples with the same duration (reshuffling) rather than longer ones (retargeting).