# I'm sorry Dave, I'm afraid I can't do that, Deep Q-learning from forbidden action

arXiv: https://arxiv.org/abs/1910.02078 · HTML: https://mlsanity.com/paper/1910.02078/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Oct 2019.

## Built on by (most cited first, 4 of 4)

- [Learning Natural Language Generation from Scratch](https://mlsanity.com/paper/2109.09371/) (2021)
- [Using Offline Data to Speed Up Reinforcement Learning in Procedurally Generated Environments](https://mlsanity.com/paper/2304.09825/) (2023)
- [Safety through Permissibility: Shield Construction for Fast and Safe Reinforcement Learning](https://mlsanity.com/paper/2405.19414/) (2024)
- [Linear Bandits beyond Inner Product Spaces, the case of Bandit Optimal Transport](https://mlsanity.com/paper/2502.07397/) (2025)

## Abstract (arXiv)

The use of Reinforcement Learning (RL) is still restricted to simulation or to enhance human-operated systems through recommendations. Real-world environments (e.g. industrial robots or power grids) are generally designed with safety constraints in mind implemented in the shape of valid actions masks or contingency controllers. For example, the range of motion and the angles of the motors of a robot can be limited to physical boundaries. Violating constraints thus results in rejected actions or entering in a safe mode driven by an external controller, making RL agents incapable of learning from their mistakes. In this paper, we propose a simple modification of a state-of-the-art deep RL algorithm (DQN), enabling learning from forbidden actions. To do so, the standard Q-learning update is enhanced with an extra safety loss inspired by structured classification. We empirically show that it reduces the number of hit constraints during the learning phase and accelerates convergence to near-optimal policies compared to using standard DQN. Experiments are done on a Visual Grid World Environment and Text-World domain.