Data-Driven Subsampling in the Presence of an Adversarial Actor (This paper accepted in IEEE ICMLCN 2024)

RML22 Dataset direct download https://drive.google.com/file/d/1E6PmQJosuKNkuv_xP8ilY_F6JP1DGdL4/view?usp=sharing

To generate the attack, we used IBM's adversarial robustness toolbox module. To run full experiment, there are four steps, 1. "ranker", 2. "subsampling", 3. "generate_adversarial_example", and 4."evaluate_attacks"