期刊
出版社
IEEE COMPUTER SOC
DOI: 10.1109/DSN.2019.00026
关键词
neural networks; classification; safety; anomaly detection
资金
- Research Grants Council of the Hong Kong Special Administrative Region, China
- CUHK [14210717]
The exceptional performance of Deep neural networks (DNNs) encourages their deployment in safety- and dependability-critical systems. However, DNNs often demonstrate erroneous behaviors in real-world corner cases. Existing counter-measures center on improving the testing and bug-fixing practice. Unfortunately, building a bug-free DNN-based system is almost impossible currently due to its black-box nature, so anomaly detection is imperative in practice. Motivated by the idea of data validation in a traditional program, we propose and implement Deep Validation, a novel framework for detecting real-world error-inducing corner cases in a DNN-based system during runtime. We model the specifications of DNNs by resorting to their training data and cast checking input validity of DNNs as the problem of discrepancy estimation. Deep Validation achieves excellent detection results against various corner case scenarios across three popular datasets. Consequently, Deep Validation greatly complements existing efforts and is a crucial step toward building safe and dependable DNN-based systems.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据