We would like to congratulate our colleagues at Heriot-Watt, Liverpool University and DSTL, namely Zhao, X., Huang, W., Banks, A., Cox, V., Flynn, D., Schewe, S., & Huang, X, on winning the best paper award at AI Safety 2021 for our research on: Assessing the Reliability of Deep Learning Classifiers Through Robustness Evaluation and Operational Profiles
This publication is concerned with a model-agnostic reliability assessment method for DL classifiers, based on evidence from robustness evaluation and the operational profile (OP) of a given application. We partition the input space into small cells and then “assemble” their robustness (to the ground truth) according to the OP, where estimators on the cells robustness and OPs are provided. A prototype tool is demonstrated with simplified case studies. While our model easily uncovers the inherent difficulties of assessing the DL dependability (e.g. lack of data with ground truth and scalability issues), we provide preliminary/compromised solutions to advance in this research direction.