Themis: Automatic and efficient deep learning system testing with strong fault detection capability

• Main Authors: HUANG, Dong, LI, Tsz On, XIE, Xiaofei, CUI, Heming
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2024
• Subjects: Deep Learning Systems (DLSs); Safety-critical tasks; Autopilot; Fault detection; Perturbed input; DLS testing techniques; Data flows; Artificial Intelligence and Robotics; Databases and Information Systems
• Online Access: https://ink.library.smu.edu.sg/sis_research/9509; https://ink.library.smu.edu.sg/context/sis_research/article/10509/viewcontent/Themis_Automatic_and_Efficient_Deep_Learning_Syste.pdf
• Institution: Singapore Management University

Deep Learning Systems (DLSs) have been widely applied in safety-critical tasks such as autopilot. However, when a perturbed input is fed into a DLS for inference, the DLS often has incorrect outputs (i.e., faults). DLS testing techniques (e.g., DeepXplore) detect such faults by generating perturbed...