Evaluating generation of chaotic time series by convolutional generative adversarial networks

Yuki Tanaka, Yutaka Yamaguti

To understand the ability and limitations of convolutional neural networks to generate time series that mimic complex temporal signals, we trained a generative adversarial network…