# Learning-based attacks in Cyber-Physical Systems: Exploration, Detection, and Control Cost trade-offs

arXiv: https://arxiv.org/abs/2011.10718 · HTML: https://mlsanity.com/paper/2011.10718/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Nov 2020.

## Built on by (most cited first, 5 of 5)

- [Understanding the Limits of Poisoning Attacks in Episodic Reinforcement Learning](https://mlsanity.com/paper/2208.13663/) (2022)
- [Saving Stochastic Bandits from Poisoning Attacks via Limited Data Verification](https://mlsanity.com/paper/2102.07711/) (2021)
- [Learning-Based Vulnerability Analysis of Cyber-Physical Systems](https://mlsanity.com/paper/2103.06271/) (2021)
- [A Secure Learning Control Strategy via Dynamic Camouflaging for Unknown Dynamical Systems under Attacks](https://mlsanity.com/paper/2102.00573/) (2021)
- [Fundamental Limits of Man-in-the-Middle Attack Detection in Model-Free Reinforcement Learning](https://mlsanity.com/paper/2603.27592/) (2026)

## Abstract (arXiv)

We study the problem of learning-based attacks in linear systems, where the communication channel between the controller and the plant can be hijacked by a malicious attacker. We assume the attacker learns the dynamics of the system from observations, then overrides the controller's actuation signal, while mimicking legitimate operation by providing fictitious sensor readings to the controller. On the other hand, the controller is on a lookout to detect the presence of the attacker and tries to enhance the detection performance by carefully crafting its control signals. We study the trade-offs between the information acquired by the attacker from observations, the detection capabilities of the controller, and the control cost. Specifically, we provide tight upper and lower bounds on the expected ε-deception time, namely the time required by the controller to make a decision regarding the presence of an attacker with confidence at least (1-εlog(1/ε)). We then show a probabilistic lower bound on the time that must be spent by the attacker learning the system, in order for the controller to have a given expected ε-deception time. We show that this bound is also order optimal, in the sense that if the attacker satisfies it, then there exists a learning algorithm with the given order expected deception time. Finally, we show a lower bound on the expected energy expenditure required to guarantee detection with confidence at least 1-ε log(1/ε).