# Don't do it: Safer Reinforcement Learning With Rule-based Guidance

arXiv: https://arxiv.org/abs/2212.13819 · HTML: https://mlsanity.com/paper/2212.13819/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 29 references that are in the map.
- First posted on arXiv 28 Dec 2022.

## Builds on (most cited first, 11 of 11)

- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [A Lyapunov-based Approach to Safe Reinforcement Learning](https://mlsanity.com/paper/1805.07708/) (2018)
- [A Review of Safe Reinforcement Learning: Methods, Theory and Applications](https://mlsanity.com/paper/2205.10330/) (2022)
- [Unsupervised State Representation Learning in Atari](https://mlsanity.com/paper/1906.08226/) (2019)
- [Trial without Error: Towards Safe Reinforcement Learning via Human Intervention](https://mlsanity.com/paper/1707.05173/) (2017)
- [Safe Reinforcement Learning with Model Uncertainty Estimates](https://mlsanity.com/paper/1810.08700/) (2018)
- [Safe Reinforcement Learning by Imagining the Near Future](https://mlsanity.com/paper/2202.07789/) (2022)
- [Safe Policy Improvement by Minimizing Robust Baseline Regret](https://mlsanity.com/paper/1607.03842/) (2016)
- [Safe Exploration of State and Action Spaces in Reinforcement Learning](https://mlsanity.com/paper/1402.0560/) (2014)
- [RAIL: Risk-Averse Imitation Learning](https://mlsanity.com/paper/1707.06658/) (2017)

## Abstract (arXiv)

During training, reinforcement learning systems interact with the world without considering the safety of their actions. When deployed into the real world, such systems can be dangerous and cause harm to their surroundings. Often, dangerous situations can be mitigated by defining a set of rules that the system should not violate under any conditions. For example, in robot navigation, one safety rule would be to avoid colliding with surrounding objects and people. In this work, we define safety rules in terms of the relationships between the agent and objects and use them to prevent reinforcement learning systems from performing potentially harmful actions. We propose a new safe epsilon-greedy algorithm that uses safety rules to override agents' actions if they are considered to be unsafe. In our experiments, we show that a safe epsilon-greedy policy significantly increases the safety of the agent during training, improves the learning efficiency resulting in much faster convergence, and achieves better performance than the base model.