Preprint · 2025 · Open access
Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards
arXiv (Cornell University) · Published 3 April 2025 · DOI 10.48550/arxiv.2504.03040
In brief
“Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards” is a 2025 preprint by Hanping Zhang and Guo, Yuhong, published in arXiv (Cornell University). OpenAlex counts 1 citations as of 22 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 1
- Most cited in
- 2026 (1)
- References
- 21
- Authors
- 2
Abstract
Safe Reinforcement Learning (Safe RL) aims to train an RL agent to maximize its performance in real-world environments while adhering to safety constraints, as exceeding safety violation limits can result in severe consequences. In this paper, we propose a novel safe RL approach called Safety Modulated Policy Optimization (SMPO), which enables safe policy function learning within the standard policy optimization framework through safety modulated rewards. In particular, we consider safety violation costs as feedback from the RL environments that are parallel to the standard awards, and introduce a Q-cost function as safety critic to estimate expected future cumulative costs. Then we propose to modulate the rewards using a cost-aware weighting function, which is carefully designed to ensure the safety limits based on the estimation of the safety critic, while maximizing the expected rewards. The policy function and the safety critic are simultaneously learned through gradient descent during online interactions with the environment. We conduct experiments using multiple RL environments and the experimental results demonstrate that our method outperforms several classic and state-of-the-art comparison methods in terms of overall safe RL performance.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Zhang, H., & Yuhong, G. (2025). Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2504.03040
MLA 9
Zhang, Hanping, and Guo, Yuhong. “Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2504.03040.
Chicago
Zhang, Hanping and Guo, Yuhong. 2025. “Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2504.03040.
Harvard
Zhang, H. and Yuhong, G. (2025) ‘Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards’, arXiv (Cornell University). doi:10.48550/arxiv.2504.03040.
IEEE
H. Zhang and G. Yuhong, “Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2504.03040.
BibTeX
@misc{zhang2025safety,
title = {{Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards}},
author = {Zhang, Hanping and Yuhong, Guo,},
howpublished = {arXiv (Cornell University)},
year = {2025},
doi = {10.48550/arxiv.2504.03040},
url = {https://doi.org/10.48550/arxiv.2504.03040}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- safe reinforcement learning
- policy optimization
- safety constraints
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Zhang, H., & Yuhong, G. (2025). Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2504.03040
How many times has this paper been cited?
1 times, according to OpenAlex on 22 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Hanping Zhang, Guo, Yuhong.
What is the DOI of this paper?
10.48550/arxiv.2504.03040, permanently at https://doi.org/10.48550/arxiv.2504.03040. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 22 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.