mlsanityData to 7 Oct 2026

Revisiting PGD Attacks for Stability Analysis of Large-Scale Nonlinear Systems and Perception-Based Control

arXiv 2201.00801 ↗First posted 3 Jan 2022math.OC cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Many existing region-of-attraction (ROA) analysis tools find difficulty in addressing feedback systems with large-scale neural network (NN) policies and/or high-dimensional sensing modalities such as cameras. In this paper, we tailor the projected gradient descent (PGD) attack method developed in the adversarial learning community as a general-purpose ROA analysis tool for large-scale nonlinear systems and end-to-end perception-based control. We show that the ROA analysis can be approximated as a constrained maximization problem whose goal is to find the worst-case initial condition which shifts the terminal state the most. Then we present two PGD-based iterative methods which can be used to solve the resultant constrained maximization problem. Our analysis is not based on Lyapunov theory, and hence requires minimum information of the problem structures. In the model-based setting, we show that the PGD updates can be efficiently performed using back-propagation. In the model-free setting (which is more relevant to ROA analysis of perception-based control), we propose a finite-difference PGD estimate which is general and only requires a black-box simulator for generating the trajectories of the closed-loop system given any initial state. We demonstrate the scalability and generality of our analysis tool on several numerical examples with large-scale NN policies and high-dimensional image observations. We believe that our proposed analysis serves as a meaningful initial step toward further understanding of closed-loop stability of large-scale nonlinear systems and perception-based control.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 40 references that are in the map. First posted on arXiv 3 Jan 2022.

Builds on

34 of 40 references.

References · 34

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2015Continuous control with deep reinforcement learningcited by 3,279
  6. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 1,916
  7. 2016Adversarial examples in the physical worldcited by 1,762
  8. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,598
  9. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,179
  10. 2017Robust Adversarial Reinforcement Learningcited by 439
  11. 2020dm_control: Software and Tasks for Continuous Controlcited by 368
  12. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 359
22 more
  1. 2019Improving Sample Efficiency in Model-Free Reinforcement Learning from Imagescited by 268
  2. 2019Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Modelcited by 228
  3. 2020Neural Lyapunov Controlcited by 142
  4. 2018The Lyapunov Neural Network: Adaptive Stability Certification for Safe Learning of Dynamical Systemscited by 87
  5. 2017Maximum Principle Based Algorithms for Deep Learningcited by 77
  6. 2021Lyapunov-stable neural-network controlcited by 66
  7. 2020Reach-SDP: Reachability Analysis of Closed-Loop Systems with Neural Network Controllers via Semidefinite Programmingcited by 43
  8. 2020Stability Analysis using Quadratic Constraints for Systems with Neural Network Controllerscited by 42
  9. 2020Enforcing robust control guarantees within neural network policiescited by 40
  10. 2019Policy Optimization for H2 Linear Control with H_∞ Robustness Guarantee: Implicit Regularization and Global Convergencecited by 39
  11. 2020Neural Certificates for Safe Control Policiescited by 35
  12. 2018Stability-certified reinforcement learning: A control-theoretic perspectivecited by 29
  13. 2018Learning Control Lyapunov Functions from Counterexamples and Demonstrationscited by 19
  14. 2021Learning Region of Attraction for Nonlinear Systemscited by 16
  15. 2021Derivative-Free Policy Optimization for Linear Risk-Sensitive and Robust Control Design: Implicit Regularization and Sample Complexitycited by 15
  16. 2017Learning Visual Servoing with Deep Features and Fitted Q-Iterationcited by 7
  17. 2019H_∞ Model-free Reinforcement Learning with Robust Stability Guaranteecited by 7
  18. 2020Learning Lyapunov Functions for Piecewise Affine Systems with Neural Network Controllerscited by 7
  19. 2018Data Driven Stability Analysis of Black-box Switched Linear Systemscited by 5
  20. 2021Model-Free μ Synthesis via Adversarial Reinforcement Learningcited by 5
  21. 2021How Are Learned Perception-Based Controllers Impacted by the Limits of Robust Control?cited by 4
  22. 2020Stability Analysis of Complementarity Systems with Neural Network Controllerscited by 2

Built on by

Not cited yet.