Using Action-Policy Testing in RL to Reduce the Number of Bugs
Open Access
- 19 July 2025
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the International Symposium on Combinatorial Search
- Vol. 18, 181-185
Abstract
Reinforcement learning is becoming ever more prominent in solving combinatorial search problems, in particular ones where states are images. Prior work has devised action-policy testing methodology, that identifies so-called bug states where policy performance is sub-optimal. Here we show how to leverage this methodology during the RL process, using action-policy testing to find bugs and injecting those as alternate start states for the training runs. Running experiments across six 2D games, we find that our testing-guided training often achieves similar expected reward while reducing the number of bugs.Keywords
This publication has 21 references indexed in Scilit:
- Safe Reinforcement Learning Through Regret and State Restorations in Evaluation Stages2024
- DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning – Extended VersionACM Transactions on Modeling and Computer Simulation, 2023
- Metamorphic relations via relaxations: an approach to obtain oracles for action-policy testing2022
- Learning Generalized Relational Heuristic Networks for Model-Agnostic PlanningProceedings of the AAAI Conference on Artificial Intelligence, 2021
- A general reinforcement learning algorithm that masters chess, shogi, and Go through self-playScience, 2018
- Falsification of Cyber-Physical Systems Using Deep Reinforcement Learning2018
- Learning Generalized Reactive Policies Using Deep Neural NetworksProceedings of the International Conference on Automated Planning and Scheduling, 2018
- Adaptive Stress Testing for Autonomous Vehicles2018
- Mastering the game of Go with deep neural networks and tree searchNature, 2016
- Efficient Guiding Strategies for Testing of Temporal Properties of Hybrid Systems2015