Adversarial Framework with Certified Robustness for Time-Series Domain via Statistical Features

DOI10.1613/JAIR.1.13543MaRDI QIDQ5076368zbMATH OpenOpenAlexFDO

Authors Taha Belkhouja, Janardhan Rao Doppa

Publication date 16 May 2022

Published in Journal of Artificial Intelligence Research (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2207.04307

zbMATH Keywords

machine learningneural networks

Mathematics Subject Classification ID

Artificial intelligence (68Txx)

Recommendations

• Masking and purifying inputs for blocking textual adversarial attacks
• Adversarial defense via the data-dependent activation, total variation minimization, and adversarial training
• A robust generative classifier against transfer attacks based on variational auto-encoders
• Detection of iterative adversarial attacks via counter attack
• Generalizing universal adversarial perturbations for deep neural networks

This page was built for publication: Adversarial Framework with Certified Robustness for Time-Series Domain via Statistical Features

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5076368)