TSFool is a multi-objective gray-box attack method to craft highly-imperceptible adversarial samples for RNN-based time series classification. The Multi-TSFool method in this repository is built for multivariate time series data, which is an extended version of the basic TSFool for univariate time series data here.
FlaAI/Multi-TSFool
Repository of the Multi-TSFool method proposed in paper "TSFool: Crafting Highly-Imperceptible Adversarial Time Series through Multi-Objective Attack".
PythonMIT