DistXplore: Distribution-guided testing for evaluating and enhancing deep learning systems

• Main Authors: WANG, Longtian, XIE, Xiaofei, DU, Xiaoning, TIAN, Meng, GUO, Qing, YANG, Zheng, SHEN, Chao
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2023
• Subjects: Deep learning; Neural networks; Software defect analysis; Software testing and debugging; Artificial Intelligence and Robotics
• Online Access: https://ink.library.smu.edu.sg/sis_research/8516; https://ink.library.smu.edu.sg/context/sis_research/article/9519/viewcontent/3611643.3616266.pdf
• Institution: Singapore Management University

Deep learning (DL) models are trained on sampled data, where the distribution of training data differs from that of real-world data (i.e., the distribution shift), which reduces the model's robustness. Various testing techniques have been proposed, including distribution-unaware and distributio...