# On the Definition of Robustness and Resilience of AI Agents for Real-time Congestion Management

arXiv: https://arxiv.org/abs/2504.13314 · HTML: https://mlsanity.com/paper/2504.13314/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 19 references that are in the map.
- First posted on arXiv 17 Apr 2025.

## Builds on (most cited first, 8 of 8)

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Challenges and Countermeasures for Adversarial Attacks on Deep Reinforcement Learning](https://mlsanity.com/paper/2001.09684/) (2020)
- [Learning to run a Power Network Challenge: a Retrospective Analysis](https://mlsanity.com/paper/2103.03104/) (2021)
- [Whatever Does Not Kill Deep Reinforcement Learning, Makes It Stronger](https://mlsanity.com/paper/1712.09344/) (2017)
- [Verification of Neural Network Behaviour: Formal Guarantees for Power System Applications](https://mlsanity.com/paper/1910.01624/) (2019)
- [Managing power grids through topology actions: A comparative study between advanced rule-based and reinforcement learning agents](https://mlsanity.com/paper/2304.00765/) (2023)
- [Understanding the Safety Requirements for Learning-based Power Systems Operations](https://mlsanity.com/paper/2110.04983/) (2021)

## Abstract (arXiv)

The European Union's Artificial Intelligence (AI) Act defines robustness, resilience, and security requirements for high-risk sectors but lacks detailed methodologies for assessment. This paper introduces a novel framework for quantitatively evaluating the robustness and resilience of reinforcement learning agents in congestion management. Using the AI-friendly digital environment Grid2Op, perturbation agents simulate natural and adversarial disruptions by perturbing the input of AI systems without altering the actual state of the environment, enabling the assessment of AI performance under various scenarios. Robustness is measured through stability and reward impact metrics, while resilience quantifies recovery from performance degradation. The results demonstrate the framework's effectiveness in identifying vulnerabilities and improving AI robustness and resilience for critical applications.