[2310.01802] A Unifying Perspective for Safety of Stochastic Systems: From Barrier Functions to Finite Abstractions

Providing safety guarantees for stochastic dynamical systems is a central problem in various fields, including control theory, machine learning, and robotics. Existing methods either employ Stochastic Barrier Functions (SBFs) or rely on numerical approaches based on finite abstractions. SBFs, analogous to Lyapunov functions, are used to establish (probabilistic) set invariance, whereas abstraction-based approaches approximate the stochastic system with a finite model to compute safety probability bounds. This paper presents a unifying perspective on these seemingly different approaches. Specifically, we show that both methods can be interpreted as approximations of a stochastic dynamic programming problem. This perspective allows us to formally establish the correctness of both techniques, characterize their convergence and optimality properties, and analyze their respective assumptions, advantages, and limitations. Our analysis reveals that, unlike SBFs-based methods, abstraction-based approaches can provide asymptotically optimal safety certificates, albeit at the cost of increased computational effort.

View PDF HTML (experimental) Abstract:Providing safety guarantees for stochastic dynamical systems is a central problem in various fields, including control theory, machine learning, and robotics. Existing methods either employ Stochastic Barrier Functions (SBFs) or rely on numerical approaches based on finite abstractions. SBFs, analogous to Lyapunov functions, are used to establish (probabilistic) set invariance, whereas abstraction-based approaches approximate the stochastic system with a finite model to compute safety probability bounds. This paper presents a unifying perspective on…

saved by

• Gilbert Chang

related reading

• Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewarxiv.org
• Robust Regression for Safe Exploration in Controlarxiv.org
• [2202.11762] Safe Control with Learned Certificates: A Survey of Neural Lyapunov, Barrier, and Contraction Methods for Robotics and Controlar5iv.labs.arxiv.org
• The Concentration of Invariant Measures for Stochastic Dynamical Systems with Locally Lipschitz Continuous Coefficients in R^darxiv.org
• ProbabilisticRobotics.pdfdocs.ufpr.br
• Generalization at the Edge of Stabilityarxiv.org
• A Review of Modern Stochastic Modeling: SDE/SPDE Numerics, Data-Driven Identification, and Generative Methods with Applications in Biology and Epidemiologyarxiv.org
• [2505.02621] Mirror Mean-Field Langevin Dynamicsarxiv.org
• An optimization perspective on log-concave sampling and beyond | Sinho Chewichewisinho.github.io
• [2605.15522] Stochastic Non-Smooth Convex Optimization with Unbounded Gradientsarxiv.org
• [2602.17616] Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMsarxiv.org
• Scalable Gradients for Stochastic Differential Equationsarxiv.org