
人工智能的未来:
负责任和可信赖
刊物
过去几年,我们在软件工程和安全领域的国际旗舰会议和期刊上发表了诸多关于人工智能检测的论文,例如ICSE、USENIX Security、CAV、TSE和FSE。
这些论文涵盖了可信赖人工智能的各个领域,包括鲁棒性、公平性、安全性和可解释性等。此外,我们还获得了两项 ACM SIGSOFT 杰出论文奖(ICSE 2018、ICSE 2020)和一项 ACM SIGSOFT 研究亮点奖(2020)。
Certified Robust Accuracy of Neural Networks Are Bounded due to Bayes Errors,CAV,2024 年 7 月
QuoTe: Quality-oriented Testing for Deep Learning Systems,TOSEM,2023 年 2 月
Adversarial Attacks and Mitigation for Anomaly Detectors of Cyber-Physical Systems,IJCIP,2021 年 9 月
Adversarial Sample Detection for Deep Neural Network through Model Mutation Testing,ICSE,2019年5月
Towards Optimal Concolic Testing,ICSE,2018年5月
鲁棒性
公平性
TestSGD: Interpretable Testing of Neural Networks Against Subtle Group Discrimination,TOSEM,2023年4月
Adaptive Fairness Improvement based Causality Analysis,FSE,2022年11月
Probabilistic Verification of Neural Networks Against Group Fairness,FM,2021年11月
Automatic Fairness Testing of Neural Classifiers Through Adversarial Sampling,TSE,2021 年 8 月
White-box Fairness Testing through Adversarial Sampling,ICSE,2020年6月
安全性
可解释性
Semantic-based Neural Network Repair,ISSTA,2023年7月
查看>>
Which Neural Network Makes More Explainable Decisions? An Approach towards Measuring Explainability,ASE-J,2022 年 11 月
ExAIs: Executable AI Semantics,ICSE,2022年7月
Towards Interpreting Recurrent Neural Network through Probabilistic Abstraction,ASE,2020 年 9 月