mlsanityData to 7 Oct 2026

Safety Filters for Black-Box Dynamical Systems by Learning Discriminating Hyperplanes

arXiv 2402.05279 ↗First posted 7 Feb 2024cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

Learning-based approaches are emerging as an effective approach for safety filters for black-box dynamical systems. Existing methods have relied on certificate functions like Control Barrier Functions (CBFs) and Hamilton-Jacobi (HJ) reachability value functions. The primary motivation for our work is the recognition that ultimately, enforcing the safety constraint as a control input constraint at each state is what matters. By focusing on this constraint, we can eliminate dependence on any specific certificate function-based design. To achieve this, we define a discriminating hyperplane that shapes the half-space constraint on control input at each state, serving as a sufficient condition for safety. This concept not only generalizes over traditional safety methods but also simplifies safety filter design by eliminating dependence on specific certificate functions. We present two strategies to learn the discriminating hyperplane: (a) a supervised learning approach, using pre-verified control invariant sets for labeling, and (b) a reinforcement learning (RL) approach, which does not require such labels. The main advantage of our method, unlike conventional safe RL approaches, is the separation of performance and safety. This offers a reusable safety filter for learning new tasks, avoiding the need to retrain from scratch. As such, we believe that the new notion of the discriminating hyperplane offers a more generalizable direction towards designing safety filters, encompassing and extending existing certificate-function-based or safe RL methodologies.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 34 of its 51 references that are in the map. First posted on arXiv 7 Feb 2024.

Builds on

34 of 51 references.

References · 34

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2017Constrained Policy Optimizationcited by 928
  4. 2021Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learningcited by 337
  5. 2018A Lyapunov-based Approach to Safe Reinforcement Learningcited by 278
  6. 2016Control Barrier Function Based Quadratic Programs for Safety Critical Systemscited by 261
  7. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 256
  8. 2020Recovery RL: Safe Reinforcement Learning with Learned Recovery Zonescited by 156
  9. 2017A General Safety Framework for Learning-Based Control in Uncertain Robotic Systemscited by 150
  10. 2022Safe Control with Learned Certificates: A Survey of Neural Lyapunov, Barrier, and Contraction methodscited by 148
  11. 2020Learning to be Safe: Deep RL with a Safety Criticcited by 109
  12. 2018A predictive safety filter for learning-based control of constrained nonlinear dynamical systemscited by 91
22 more
  1. 2020DeepReach: A Deep Learning Approach to High-Dimensional Reachabilitycited by 81
  2. 2019Learning for Safety-Critical Control with Control Barrier Functionscited by 66
  3. 2020Reinforcement Learning for Safety-Critical Control under Model Uncertainty, using Control Lyapunov Functions and Control Barrier Functionscited by 61
  4. 2021Safe Reinforcement Learning Using Advantage-Based Interventioncited by 45
  5. 2021Safe Reinforcement Learning Using Robust Control Barrier Functionscited by 44
  6. 2023How to Train Your Neural Control Barrier Function: Learning Safety Filters for Complex Input-Constrained Systemscited by 40
  7. 2023Not Only Rewards But Also Constraints: Applications on Legged Robot Locomotioncited by 33
  8. 2022Safe Control Under Input Limits with Neural Control Barrier Functionscited by 27
  9. 2022ISAACS: Iterative Soft Adversarial Actor-Critic for Safetycited by 24
  10. 2023Trust Region-Based Safe Distributional Reinforcement Learning for Multiple Constraintscited by 20
  11. 2021Learning Robust Output Control Barrier Functions from Safe Expert Demonstrationscited by 19
  12. 2019Safe Control Algorithms Using Energy Functions: A Unified Framework, Benchmark, and New Directionscited by 17
  13. 2020Control Barriers in Bayesian Learning of System Dynamicscited by 17
  14. 2023In-Distribution Barrier Functions: Self-Supervised Policy Filters that Avoid Out-of-Distribution Statescited by 15
  15. 2020Towards Robust Data-Driven Control Synthesis for Nonlinear Systems with Actuation Uncertaintycited by 13
  16. 2021Pointwise Feasibility of Gaussian Process-based Safety-Critical Control under Model Uncertaintycited by 13
  17. 2020Safe reinforcement learning for probabilistic reachability and safety specifications: A Lyapunov-based approachcited by 10
  18. 2022Barrier Bayesian Linear Regression: Online Learning of Control Barrier Conditions for Safety-Critical Control of Uncertain Systemscited by 4
  19. 2023A Forward Reachability Perspective on Control Barrier Functions and Discount Factors in Reachability Analysiscited by 3
  20. 2017Scalable synthesis of safety certificates from data with application to learning-based controlcited by 2
  21. 2022Reach-avoid Verification Based on Convex Optimizationcited by 2
  22. 2023Uncertainty Quantification for Recursive Estimation in Adaptive Safety-Critical Controlcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 2; 2026: 3.

8 papers.

  1. 2024Safe Reinforcement Learning on the Constraint Manifold: Theory and Applicationscited by 12
  2. 2025Reachability Barrier Networks: Learning Hamilton-Jacobi Solutions for Smooth and Flexible Control Barrier Functionscited by 3
  3. 2024Learning Ensembles of Vision-based Safety Control Filterscited by 2
  4. 2025A Simple Approach to Constraint-Aware Imitation Learning with Application to Autonomous Racingcited by 2
  5. 2024Pre-Trained Vision Models as Perception Backbones for Safety Filters in Autonomous Drivingcited by 1
  6. 2026Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertaintycited by 1
  7. 2026CAffNet: Hard Constraint-Affine Neural Networkscited by 0
  8. 2026All Models are Wrong, Knowing Where is Useful: On Model Uncertainty in Reinforcement Learningcited by 0