ConAML: Constrained Adversarial Machine Learning for Cyber-Physical Systems
This addresses security risks for cyber-physical systems like power grids and water treatment, but it is incremental as it adapts existing adversarial machine learning methods to incorporate physical constraints.
The authors tackled the vulnerability of machine learning models in cyber-physical systems to adversarial examples by proposing ConAML, which generates adversarial examples that satisfy physical constraints, and demonstrated its effectiveness in simulations of power grids and water treatment systems, significantly decreasing model performance.
Recent research demonstrated that the superficially well-trained machine learning (ML) models are highly vulnerable to adversarial examples. As ML techniques are becoming a popular solution for cyber-physical systems (CPSs) applications in research literatures, the security of these applications is of concern. However, current studies on adversarial machine learning (AML) mainly focus on pure cyberspace domains. The risks the adversarial examples can bring to the CPS applications have not been well investigated. In particular, due to the distributed property of data sources and the inherent physical constraints imposed by CPSs, the widely-used threat models and the state-of-the-art AML algorithms in previous cyberspace research become infeasible. We study the potential vulnerabilities of ML applied in CPSs by proposing Constrained Adversarial Machine Learning (ConAML), which generates adversarial examples that satisfy the intrinsic constraints of the physical systems. We first summarize the difference between AML in CPSs and AML in existing cyberspace systems and propose a general threat model for ConAML. We then design a best-effort search algorithm to iteratively generate adversarial examples with linear physical constraints. We evaluate our algorithms with simulations of two typical CPSs, the power grids and the water treatment system. The results show that our ConAML algorithms can effectively generate adversarial examples which significantly decrease the performance of the ML models even under practical constraints.