Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 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 Generalizable Physics-informed Learning Framework for Risk Probability Estimation (2305.06432v3)

Published 10 May 2023 in eess.SY, cs.LG, and cs.SY

Abstract: Accurate estimates of long-term risk probabilities and their gradients are critical for many stochastic safe control methods. However, computing such risk probabilities in real-time and in unseen or changing environments is challenging. Monte Carlo (MC) methods cannot accurately evaluate the probabilities and their gradients as an infinitesimal devisor can amplify the sampling noise. In this paper, we develop an efficient method to evaluate the probabilities of long-term risk and their gradients. The proposed method exploits the fact that long-term risk probability satisfies certain partial differential equations (PDEs), which characterize the neighboring relations between the probabilities, to integrate MC methods and physics-informed neural networks. We provide theoretical guarantees of the estimation error given certain choices of training configurations. Numerical results show the proposed method has better sample efficiency, generalizes well to unseen regions, and can adapt to systems with changing parameters. The proposed method can also accurately estimate the gradients of risk probabilities, which enables first- and second-order techniques on risk probabilities to be used for learning and control.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (40)
  1. Tensorflow: Large-scale machine learning on heterogeneous distributed systems. arXiv preprint arXiv:1603.04467, 2016.
  2. Probabilistic reachability and safety for controlled discrete time stochastic hybrid systems. Automatica, 44(11):2724–2734, 2008.
  3. Reachability-based safe learning with gaussian processes. In 53rd IEEE Conference on Decision and Control, pages 1424–1431. IEEE, 2014.
  4. Control barrier functions: Theory and applications. In 2019 18th European control conference (ECC), pages 3420–3431. IEEE, 2019.
  5. Estimation of small failure probabilities in high dimensions by subset simulation. Probabilistic engineering mechanics, 16(4):263–277, 2001.
  6. Markov chain importance sampling with applications to rare event probability estimation. Statistics and Computing, 23(2):271–285, 2013.
  7. Concentration inequalities: A nonasymptotic theory of independence. Oxford university press, 2013.
  8. Physics-informed neural networks (pinns) for fluid mechanics: A review. Acta Mechanica Sinica, pages 1–12, 2022.
  9. Sequential monte carlo for rare event estimation. Statistics and computing, 22(3):795–808, 2012.
  10. A risk-sensitive finite-time reachability approach for safety of stochastic dynamic systems. In 2019 American Control Conference (ACC), pages 2958–2963. IEEE, 2019.
  11. Safe control in the presence of stochastic uncertainties. In 2021 60th IEEE Conference on Decision and Control (CDC), pages 6640–6645. IEEE, 2021.
  12. William S Cleveland. Lowess: A program for smoothing scatterplots by robust locally weighted regression. American Statistician, 35(1):54, 1981.
  13. George Cybenko. Approximation by superpositions of a sigmoidal function. Mathematics of control, signals and systems, 2(4):303–314, 1989.
  14. Safe learning of uncertain environments. arXiv preprint arXiv:2103.01413, 2021.
  15. Elliptic partial differential equations of second order, volume 224. Springer, 1977.
  16. Solving high-dimensional partial differential equations using deep learning. Proceedings of the National Academy of Sciences, 115(34):8505–8510, 2018.
  17. Hans Janssen. Monte-carlo based uncertainty analysis: Sampling efficiency and sampling convergence. Reliability Engineering & System Safety, 109:123–132, 2013.
  18. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  19. Toward safe close-proximity human-robot interaction with standard industrial robots. In 2014 IEEE International Conference on Automation Science and Engineering (CASE), pages 339–344. IEEE, 2014.
  20. An analytical safe approximation to joint chance-constrained programming with additive gaussian noises. IEEE Transactions on Automatic Control, 66(11):5490–5497, 2021.
  21. Safe exploration: Addressing various uncertainty levels in human robot interactions. In 2015 American Control Conference (ACC), pages 465–470. IEEE, 2015.
  22. Deepxde: A deep learning library for solving differential equations. SIAM Review, 63(1):208–228, 2021.
  23. Physics-informed neural networks for power systems. In 2020 IEEE Power & Energy Society General Meeting (PESGM), pages 1–5. IEEE, 2020.
  24. Chance-constrained trajectory optimization for safe exploration and learning of nonlinear systems. IEEE Robotics and Automation Letters, 6(2):389–396, 2020.
  25. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems, 32, 2019.
  26. Upper bounds for continuous-time end-to-end risks in stochastic robot navigation. In 2022 European Control Conference (ECC), pages 2049–2055. IEEE, 2022.
  27. Accelerating physics-informed neural network training with prior dictionaries. arXiv preprint arXiv:2004.08151, 2020.
  28. Safe optimal control using stochastic barrier functions and deep forward-backward sdes. In Conference on Robot Learning, pages 1783–1801. PMLR, 2021.
  29. Safe reinforcement learning with chance-constrained model predictive control. In Learning for Dynamics and Control Conference, pages 291–303. PMLR, 2022.
  30. A framework for worst-case and stochastic safety verification using barrier certificates. IEEE Transactions on Automatic Control, 52(8):1415–1428, 2007.
  31. Application of reachability analysis for stochastic hybrid systems to aircraft conflict prediction. In 2008 47th IEEE Conference on Decision and Control, pages 4036–4041. IEEE, 2008.
  32. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational physics, 378:686–707, 2019.
  33. Rare event simulation using Monte Carlo methods. John Wiley & Sons, 2009.
  34. Physics-informed neural networks for cardiac activation mapping. Frontiers in Physics, 8:42, 2020.
  35. Safety-aware optimal control of stochastic systems using conditional value-at-risk. In 2018 Annual American Control Conference (ACC), pages 6285–6290. IEEE, 2018.
  36. A barrier function approach to finite-time stochastic system verification and control. Automatica, 125:109439, 2021.
  37. Safe control for nonlinear systems with stochastic uncertainty via risk control barrier functions. arXiv preprint arXiv:2203.15892, 2022.
  38. Myopically verifiable probabilistic certificate for long-term safety. arXiv preprint arXiv:2110.13380, 2021.
  39. Chance-constrained collision avoidance for mavs in dynamic environments. IEEE Robotics and Automation Letters, 4(2):776–783, 2019.
  40. Konstantin Zuev. Subset simulation method for rare event estimation: an introduction. arXiv preprint arXiv:1505.03506, 2015.
Citations (4)

Summary

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