Self-checking deep neural networks for anomalies and adversaries in deployment

• Main Authors: XIAO, Yan, BESCHASTNIKH, Ivan, LIN, Yun, HUNDAL, Rajdeep Singh, XIE, Xiaofei, ROSENBLUM, David S., DONG, Jin Song
• 格式: text
• 語言: English
• 出版: Institutional Knowledge at Singapore Management University 2022
• 主題: self-checking system; trustworthiness; deep neural networks; adversarial examples; deployment; Artificial Intelligence and Robotics; OS and Networks
• 在線閱讀: https://ink.library.smu.edu.sg/sis_research/7493; https://ink.library.smu.edu.sg/context/sis_research/article/8496/viewcontent/tdsc22_selfchecker.pdf

Deep Neural Networks (DNNs) have been widely adopted, yet DNN models are surprisingly unreliable, which raises significant concerns about their use in critical domains. In this work, we propose that runtime DNN mistakes can be quickly detected and properly dealt with in deployment, especially in set...