Shiqi Sun
Papers
1
Total Citations
6
H-Index
1
About
Shiqi Sun is a rising researcher at the forefront of formal methods for safety-critical autonomous systems, with a primary focus on the verification of stochastic cyber-physical systems (CPS) controlled by neural networks. In their most cited work, "Formal Verification of Stochastic Systems with ReLU Neural Network Controllers" (2022, 6 citations), Sun tackles the fundamental challenge of guaranteeing safety under uncertainty. They developed a rigorous framework to compute the set of initial states from which a system, governed by both probabilistic dynamics and a ReLU neural network controller, will almost surely avoid unsafe configurations. This contribution bridges the gap between deep learning-based control and formal safety assurance, a critical step for deploying AI in high-stakes environments like autonomous driving and robotics. By integrating reachability analysis with probabilistic guarantees, Sun’s work provides a scalable verification tool that does not sacrifice mathematical rigor. Their research is particularly notable for addressing the non-convex, piecewise-linear nature of ReLU networks within a stochastic setting—a problem that has long challenged the formal verification community. As an early-career scholar, Shiqi Sun is establishing a reputation for producing foundational, theoretically sound methods that promise to make autonomous systems both smarter and safer.
Research Focus
Key Achievements
Top Papers
- 1