Inverting Self-Triggered Control: Adversarial Reinforcement Learning for Sparse Denial-of-Service Attacks

• URL: http://arxiv.org/abs/2609.12016v1
• Date: Thu, 10 Sep 2026 07:21:54 GMT
• Title: Inverting Self-Triggered Control: Adversarial Reinforcement Learning for Sparse Denial-of-Service Attacks
• Abstract summary: Self-triggered reinforcement learning control (RL-STC) learns the sparsest control schedule that preserves Lyapunov-decreasing stability.<n>We prove a plant-property lower bound on the minimum jam count required for an immediate hold-last medium-access-control adversary.
• Score: 0.27998963147546146
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 $2.8\times$, 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.

Related papers

• Agents That Certify Their Own Exploits: Confidence-Scheduled Restricted Responses for Safe Opponent Exploitation [38.06569764716213]
  We introduce emphbudget-constrained confidence-scheduled restricted responses (CS-RNR)<n>CS-RNR is the first opponent-exploitation method whose safety guarantee is a certificate the agent computes on the strategy it actually deploys.
  arXiv  Detail & Related papers  (2026-07-30T16:57:57Z)
• Real-Time Rulebook-Aware Nonlinear MPC for Autonomous Driving with Priority-Biased Tiered Slacks [0.8363171780853939]
  W-SQP is a weighted tiered-slack nonlinear model predictive controller (NMPC)<n>We present W-SQP, a weighted tiered-slack nonlinear model predictive controller (NMPC)<n>We present W-SQP, a weighted tiered-slack nonlinear model predictive controller (NMPC)
  arXiv  Detail & Related papers  (2026-07-13T00:40:27Z)
• ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples [65.74946647574696]
  Reinforcement learning (RL) has significantly enhanced the reasoning capabilities of large language models (LLMs)<n>In this work, we investigate a critical source of this instability: over-optimization.<n>We propose ARMOR, a framework that shifts the paradigm from passive penalty to active sample stabilization.
  arXiv  Detail & Related papers  (2026-07-11T21:22:46Z)
• Capacity-Constrained Online Convex Optimization with Delayed Feedback [58.59385794080679]
  We study delayed online convex optimization (OCO) under a hard capacity constraint.<n>We find that for first-order feedback, capacity $C = (log T)$ suffices to recover standard delayed OCO rates up to logarithmic factors.<n>For bandit feedback, the regret rates are modulated by powers of $(1 + _textmax/C)$, where $_textmax$ is the maximum number of pending observations at any time.
  arXiv  Detail & Related papers  (2026-06-10T06:37:06Z)
• Anytime-Valid Federated Conformal RAG for LLM Swarms [12.805268849262243]
  Anytime-FC-RAG is a sequential extension built on a summable per-step calibration-deviation budget.<n>Experiments on a GPT-2-small + MiniLM swarm across MMLU, DBpedia, and AG News verify the predicted alarm rate, detection delay, envelope coverage, and $14$-$57%$ bandwidth savings.
  arXiv  Detail & Related papers  (2026-05-27T22:06:57Z)
• Survive or Collapse: The Asymmetric Roles of Data Gating and Reward Grounding in Self-Play RL [76.45061154544568]
  Self-play reinforcement learning trains language models on their own generated tasks, co-evolving a proposer and solver without human labels.<n>Recent systems report strong reasoning gains, but collapse and instability are widely observed and poorly understood.<n>We argue instead that self-play stability is governed by two distinct levers: a data-level gate that decides which proposer-generated tasks enter the training pool, and the reward signal that updates the policy on tasks already admitted.
  arXiv  Detail & Related papers  (2026-05-21T09:19:23Z)
• Learning When to Act: Communication-Efficient Reinforcement Learning via Run-Time Assurance [0.27998963147546146]
  We show that a single policy can learn control inputs and communication-efficient timing decisions under a pointwise Lyapunov safety shield.<n>A run-time assurance layer overrides the policy via a one-step-ahead Lyapunov prediction and a precomputed LQR backup.<n>A 12-state 3D quadrotor case study extends the framework to higher-dimensional systems where classical STC is intractable.
  arXiv  Detail & Related papers  (2026-05-11T23:55:15Z)
• What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time [57.533031432715084]
  Test-Time Reinforcement Learning (TTRL) enables Large Language Models (LLMs) to enhance reasoning capabilities on unlabeled test streams.<n>Existing TTRL methods rely exclusively on positive pseudo-labeling strategies.<n>We propose SCRL, a robust test-time reinforcement learning framework that effectively mitigates label noise amplification.
  arXiv  Detail & Related papers  (2026-03-20T11:47:12Z)
• TCRL: Temporal-Coupled Adversarial Training for Robust Constrained Reinforcement Learning in Worst-Case Scenarios [7.3881187966342905]
  Constrained Reinforcement Learning aims to optimize decision-making policies under constraint conditions.<n>Existing robust CRL approaches predominantly focus on single-step perturbations and temporally independent adversarial models.<n>We propose temporal-coupled adversarial training framework for robust constrained reinforcement learning (TCRL) in worst-case scenarios.
  arXiv  Detail & Related papers  (2026-02-13T15:48:20Z)
• Time Is All It Takes: Spike-Retiming Attacks on Event-Driven Spiking Neural Networks [87.16809558673403]
  Spiking neural networks (SNNs) compute with discrete spikes and exploit temporal structure.<n>We study a timing-only adversary that retimes existing spikes while preserving spike counts and amplitudes in event-driven SNNs.
  arXiv  Detail & Related papers  (2026-02-03T09:06:53Z)
• Improper Learning for Non-Stochastic Control [78.65807250350755]
  We consider the problem of controlling a possibly unknown linear dynamical system with adversarial perturbations, adversarially chosen convex loss functions, and partially observed states. Applying online descent to this parametrization yields a new controller which attains sublinear regret vs. a large class of closed-loop policies. Our bounds are the first in the non-stochastic control setting that compete with emphall stabilizing linear dynamical controllers.
  arXiv  Detail & Related papers  (2020-01-25T02:12:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.