Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
184 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A Toolbox for Fast Interval Arithmetic in numpy with an Application to Formal Verification of Neural Network Controlled Systems (2306.15340v1)

Published 27 Jun 2023 in eess.SY, cs.LG, cs.SY, and math.OC

Abstract: In this paper, we present a toolbox for interval analysis in numpy, with an application to formal verification of neural network controlled systems. Using the notion of natural inclusion functions, we systematically construct interval bounds for a general class of mappings. The toolbox offers efficient computation of natural inclusion functions using compiled C code, as well as a familiar interface in numpy with its canonical features, such as n-dimensional arrays, matrix/vector operations, and vectorization. We then use this toolbox in formal verification of dynamical systems with neural network controllers, through the composition of their inclusion functions.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (15)
  1. Computing robustly forward invariant sets for mixed-monotone systems. In 2020 59th IEEE Conference on Decision and Control (CDC), pp.  4553–4559, 2020. doi:10.1109/CDC42340.2020.9304461.
  2. Tight decomposition functions for continuous-time mixed-monotone systems with disturbances. IEEE Control Systems Letters, 5(1):139–144, 2021. doi:10.1109/LCSYS.2020.3001085.
  3. Scalable verified training for provably robust image classification. In IEEE/CVF International Conference on Computer Vision (ICCV), pp.  4841–4850, 2019. doi:10.1109/ICCV.2019.00494.
  4. Array programming with NumPy. Nature, 585:357–362, 2020. doi:10.1038/s41586-020-2649-2.
  5. Interval arithmetic: From principles to implementation. J. ACM, 48(5):1038–1068, sep 2001. ISSN 0004-5411. doi:10.1145/502102.502106.
  6. POLAR: A polynomial arithmetic framework for verifying neural-network controlled systems. In Automated Technology for Verification and Analysis: 20th International Symposium, ATVA 2022, Virtual Event, October 25-28, 2022, Proceedings, pp.  414–430. Springer, 2022. doi:10.48550/arXiv.2106.13867.
  7. Interval reachability of nonlinear dynamical systems with neural network controllers. In Learning for Dynamics and Control Conference, volume 211, pp.  1–14. PMLR, 2023. doi:10.48550/arXiv.2304.03671.
  8. Applied Interval Analysis. Springer London, 2001. doi:10.1007/978-1-4471-0249-6.
  9. TIRA: Toolbox for interval reachability analysis. In Proceedings of the 22nd ACM International Conference on Hybrid Systems: Computation and Control, pp.  224–229, 2019. doi:10.1145/3302504.3311808.
  10. Stability of Dynamical Systems: Continuous, Discontinuous, and Discrete Systems. Stability of Dynamical Systems: Continuous, Discontinuous, and Discrete Systems. Birkhäuser Boston, 2008. doi:10.1007/978-0-8176-4649-3. URL https://books.google.com/books?id=s4mq1h0elUkC.
  11. The kinematic bicycle model: A consistent model for planning feasible trajectories for autonomous vehicles? In 2017 IEEE Intelligent Vehicles Symposium (IV), pp. 812–818, 2017. doi:10.1109/IVS.2017.7995816.
  12. Bounds on the reachable sets of nonlinear control systems. Automatica, 49(1):93–100, 2013. ISSN 0005-1098. doi:10.1016/j.automatica.2012.09.020.
  13. Reachable set estimation for neural network control systems: A simulation-guided approach. IEEE Transactions on Neural Networks and Learning Systems, 32(5):1821–1830, 2021. doi:10.1109/TNNLS.2020.2991090.
  14. Automatic perturbation analysis for scalable certified robustness and beyond. Advances in Neural Information Processing Systems, 33, 2020. doi:10.48550/arXiv.2002.12920.
  15. Efficient neural network robustness certification with general activation functions. In Advances in Neural Information Processing Systems, volume 31, pp.  4944–4953, 2018. doi:10.48550/arXiv.1811.00866.
Citations (11)

Summary

We haven't generated a summary for this paper yet.