@article{
daneshmand2026safeexplorer,
title={SafeExplorer: An Unbiased Mixed-Policy Gradient for Reinforcement Learning with Recovery Interventions},
author={Elham Daneshmand and Majid Khadiv and Glen Berseth and Hsiu-Chin Lin},
journal={Transactions on Machine Learning Research},
issn={2835-8856},
year={2026},
url={https://openreview.net/forum?id=ppOrbSsMM0},
note={}
}