Bridging Formal Methods and Machine Learning with Global Optimisation

Xiaowei Huang, Wenjie Ruan, Qiyi Tang, Xingyu Zhao

Abstract

Formal methods and machine learning are two research fields with drastically different foundations and philosophies. Formal methods utilise mathematically rigorous techniques for the specification, development and verification of software and hardware systems. Machine learning focuses on pragmatic approaches to gradually improve a parameterised model by observing a training data set. While historically the two fields lack communication, this trend has changed in the past few years with an outburst of research interest in the robustness verification of neural networks. This paper will briefly review these works, and focus on the urgent need for broader, and more in-depth, communication between the two fields, with the ultimate goal of developing learning-enabled systems with not only excellent performance but also acceptable safety and security. We present a specification language, MLS $$^2$$ , and show that it can express a set of known safety and security properties, including generalisation, uncertainty, robustness, data poisoning, backdoor, model stealing, membership inference, model inversion, interpretability, and fairness. To verify MLS $$^2$$ properties, we promote the global optimisation based methods, which have provable guarantees on the convergence to the optimal solution. Many of them have theoretical bounds on the gap between current solutions and the optimal solution.
BibTeX
@inproceedings{huang2022bridging,
  author = {Xiaowei Huang and Wenjie Ruan and Qiyi Tang and Xingyu Zhao},
  title = {{Bridging Formal Methods and Machine Learning with Global Optimisation}},
  booktitle = {ICFEM},
  year = {2022},
  doi = {10.1007/978-3-031-17244-1\_1},
}

Loading related content...

Figure