# Simulating SQL Injection Vulnerability Exploitation Using Q-Learning Reinforcement Learning Agents

arXiv: https://arxiv.org/abs/2101.03118 · HTML: https://mlsanity.com/paper/2101.03118/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 23 references that are in the map.
- First posted on arXiv 8 Jan 2021.

## Builds on (most cited first, 4 of 4)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Automated Vulnerability Detection in Source Code Using Deep Representation Learning](https://mlsanity.com/paper/1807.04320/) (2018)
- [Penetration Testing == POMDP Solving?](https://mlsanity.com/paper/1306.4714/) (2013)
- [The Agent Web Model -- Modelling web hacking for reinforcement learning](https://mlsanity.com/paper/2009.11274/) (2020)

## Built on by (most cited first, 6 of 6)

- [BertRLFuzzer: A BERT and Reinforcement Learning Based Fuzzer](https://mlsanity.com/paper/2305.12534/) (2023)
- [SoK: On the Offensive Potential of AI](https://mlsanity.com/paper/2412.18442/) (2024)
- [SoK: The Pitfalls of Deep Reinforcement Learning for Cybersecurity](https://mlsanity.com/paper/2602.08690/) (2026)
- [A Unified Modeling Framework for Automated Penetration Testing](https://mlsanity.com/paper/2502.11588/) (2025)
- [Responsible Development of Offensive AI](https://mlsanity.com/paper/2504.02701/) (2025)
- [Clutch Control: An Attention-based Combinatorial Bandit for Efficient Mutation in JavaScript Engine Fuzzing](https://mlsanity.com/paper/2510.12732/) (2025)

## Abstract (arXiv)

In this paper, we propose a formalization of the process of exploitation of SQL injection vulnerabilities. We consider a simplification of the dynamics of SQL injection attacks by casting this problem as a security capture-the-flag challenge. We model it as a Markov decision process, and we implement it as a reinforcement learning problem. We then deploy reinforcement learning agents tasked with learning an effective policy to perform SQL injection; we design our training in such a way that the agent learns not just a specific strategy to solve an individual challenge but a more generic policy that may be applied to perform SQL injection attacks against any system instantiated randomly by our problem generator. We analyze the results in terms of the quality of the learned policy and in terms of convergence time as a function of the complexity of the challenge and the learning agent's complexity. Our work fits in the wider research on the development of intelligent agents for autonomous penetration testing and white-hat hacking, and our results aim to contribute to understanding the potential and the limits of reinforcement learning in a security environment.