著者

Yuki Tanaka Yutaka Yamaguti

出版者

The Japan Society for Industrial and Applied Mathematics

雑誌

JSIAM Letters (ISSN:18830609)

巻号頁・発行日

vol.15, pp.117-120, 2023 (Released:2023-11-16)

参考文献数

14

---

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 consisting of convolutional networks to generate chaotic time series and used nonlinear time series analysis to evaluate the generated time series. A numerical measure of determinism and the Lyapunov exponent showed that the generated time series well reproduce the chaotic properties of the original time series. However, error distribution analyses showed that large errors appeared at a low but non-negligible rate. Such errors would not be expected if the distribution were assumed to be exponential.