Inverting Self-Triggered Control: Adversarial Reinforcement Learning for Sparse Denial-of-Service Attacks
On the map: Deep reinforcement learning methods → Adversarial robustness and RL in finance
Abstract (arXiv)
Self-triggered reinforcement learning control (RL-STC) learns the sparsest control schedule that preserves Lyapunov-decreasing stability under a Run-Time Assurance (RTA) override. We invert this: an adversarial RL agent learns the sparsest jamming or Denial-of-Service (DoS) schedule that destabilizes the closed loop, with a Lyapunov-increase admissibility predicate mirroring the defender's safety certificate. We prove a plant-property lower bound on the minimum jam count required for an immediate hold-last medium-access-control adversary to force a crash against a self-triggered controller (STC) satisfying a Lyapunov contract, and recover a certificate-level analog of the consecutive-grouping optimality of prior count-budget DoS scheduling as a corollary. This extends the DoS-scheduling count-budget analysis from periodic and linear-time-invariant to STC controllers. Empirically, we train against four fixed defenders per plant (one Linear Quadratic Regulator (LQR) and three RL-STC) on Pendulum, CartPole, and Quadrotor2D. The learned adversary is the only adversary that crashes every defender on every plant at 100%: greedy misses Quadrotor2D LQR on 42% of episodes and periodic misses Pendulum LQR on 97%. On jam-time-per-failure it beats baselines by up to 2.8×, and shows its widest absolute margin on Quadrotor2D LQR. Robustness ablations show that Gaussian observation noise exceeding the initial-state magnitude and position-only observation both preserve 100% failure rate and keep the learned adversary strictly ahead of both baselines on jam-time-per-failure.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Adversarial robustness and RL in finance, one of 15 directions in Deep reinforcement learning methods. Builds on 6 of its 17 references that are in the map. First posted on arXiv 10 Sep 2026.
Builds on
6 of 17 references.
References · 6
- 2021Jamming Attacks and Anti-Jamming Strategies in Wireless Networks: A Comprehensive Surveycited by 30
- 2021Reinforcement Learning for Feedback-Enabled Cyber Resiliencecited by 19
- 2017Analysis of Stochastic Switched Systems with Application to Networked Control Under Jamming Attackscited by 2
- 2017Stabilization of Networked Control Systems under DoS Attacks and Output Quantizationcited by 2
- 2021A Jointly Optimal Design of Control and Scheduling in Networked Systems under Denial-of-Service Attackscited by 2
- 2026Learning When to Act: Communication-Efficient Reinforcement Learning via Run-Time Assurancecited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2024Resilient Learning-Based Control Under Denial-of-Service Attacks
- 2026Taming the Adversary: A Cost-to-Disturbance Ratio Approach to Adversarial Reinforcement Learning
- 2025Provably Invincible Adversarial Attacks on Reinforcement Learning Systems: A Rate-Distortion Information-Theoretic Approach
- 2026Sparsity Is Necessary: Polynomial-Time Stability for Agentic LLMs in Large Action Spaces
- 2024Enhancing Robustness in Deep Reinforcement Learning: A Lyapunov Exponent Approach
- 2024Resilient control under denial-of-service and uncertainty: An adaptive dynamic programming approach ↗
- 2026FoSeRL: Formal Sequential Robustness Certification for Reinforcement Learning Policies
- 2024Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies