Generating Time Series by Matching Random Convolutional Features

Nikita Zozoulenko, Konrad Jakob Mueller, Thomas Cass, Lukas Gonon

Keywords: time series generation, financial time series, data scarcity, feature matching, random convolutions, signature transform, maximum mean discrepancy, mmd

TL;DR: We introduce SOCK, a differentiable convolutional feature map for training generative models via feature matching, which outperforms established methods on challenging financial time series and two-sample hypothesis testing tasks.

Abstract: Generating realistic financial time series is challenging as training data is typically limited to a single historical path, making adversarial training prone to discriminator overfitting. Instead, recent work replaces the trained discriminator with a fixed feature map, and trains to generate paths whose features match those of real time series. As features, most prior work chooses path statistics based on signatures from rough path theory. Motivated by strong empirical results of random convolutional features on time series classification, we train generative models with **SOCK** (***S**oftmax-Variance **O**f **C**ompeting **K**ernels*) a novel random convolutional feature map that is fully differentiable, highly scalable, and simple to implement. SOCK consistently outperforms other convolutional or signature-based feature maps on hypothesis testing benchmarks and on training generative models.

Submission Number: 89