Conference paper · 2024 · Open access

Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization

Homayoun Honari, Mehran Ghafarian Tamizi, Homayoun Najjaran

Proceedings - IEEE International Conference on Robotics and Automation/Proceedings, 2873–2879 · Published 13 May 2024 · DOI 10.1109/icra57147.2024.10611316

Open PDFView at the publisherCite this paper

In brief

“Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization” is a 2024 conference paper by Homayoun Honari and 2 co-authors, published in Proceedings - IEEE International Conference on Robotics and Automation/Proceedings. OpenAlex counts 11 citations as of 22 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
11
Most cited in
2026 (6)
Field-weighted impact
0.85
References
26
Authors
3

Abstract

Safe reinforcement learning (Safe RL) refers to a class of techniques that aim to prevent RL algorithms from violating constraints in the process of decision-making and exploration during trial and error. In this paper, a novel model-free Safe RL algorithm, formulated based on the multi-objective policy optimization framework is introduced where the policy is optimized towards optimality and safety, simultaneously. The optimality is achieved by the environment reward function that is subsequently shaped using a safety critic. The advantage of the Safety Optimized RL (SORL) algorithm compared to the traditional Safe RL algorithms is that it omits the need to constrain the policy search space. This allows SORL to find a natural tradeoff between safety and optimality without compromising the performance in terms of either safety or optimality due to strict search space constraints. Through our theoretical analysis of SORL, we propose a condition for SORL’s converged policy to guarantee safety and then use it to introduce an aggressiveness parameter that allows for fine-tuning the mentioned tradeoff. The experimental results obtained in seven different robotic environments indicate a considerable reduction in the number of safety violations along with higher, or competitive, policy returns, in comparison to six different state-of-the-art Safe RL methods. The results demonstrate the significant superiority of the proposed SORL algorithm in safety-critical applications.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Honari, H., Tamizi, M. G., & Najjaran, H. (2024). Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization. Proceedings - IEEE International Conference on Robotics and Automation/Proceedings, 2873–2879. https://doi.org/10.1109/icra57147.2024.10611316

MLA 9

Honari, Homayoun, et al. “Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization.” Proceedings - IEEE International Conference on Robotics and Automation/Proceedings, 2024, pp. 2873–2879. https://doi.org/10.1109/icra57147.2024.10611316.

Chicago

Honari, Homayoun, Mehran Ghafarian Tamizi, and Homayoun Najjaran. 2024. “Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization.” Proceedings - IEEE International Conference on Robotics and Automation/Proceedings: 2873–2879. https://doi.org/10.1109/icra57147.2024.10611316.

Harvard

Honari, H., Tamizi, M. G. and Najjaran, H. (2024) ‘Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization’, Proceedings - IEEE International Conference on Robotics and Automation/Proceedings, pp. 2873–2879. doi:10.1109/icra57147.2024.10611316.

IEEE

H. Honari, M. G. Tamizi, and H. Najjaran, “Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization,” Proceedings - IEEE International Conference on Robotics and Automation/Proceedings, pp. 2873–2879, 2024, doi: 10.1109/icra57147.2024.10611316.

BibTeX

@inproceedings{honari2024safety,
  title     = {{Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization}},
  author    = {Honari, Homayoun and Tamizi, Mehran Ghafarian and Najjaran, Homayoun},
  booktitle = {Proceedings - IEEE International Conference on Robotics and Automation/Proceedings},
  year      = {2024},
  pages     = {2873--2879},
  doi       = {10.1109/icra57147.2024.10611316},
  url       = {https://doi.org/10.1109/icra57147.2024.10611316}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
202411
202544
202666

Its field-weighted citation impact is 0.85: it has been cited 0.85 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics

  • safe reinforcement learning
  • safety-critical robotics
  • model-free reinforcement learning

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Honari, H., Tamizi, M. G., & Najjaran, H. (2024). Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization. Proceedings - IEEE International Conference on Robotics and Automation/Proceedings, 2873–2879. https://doi.org/10.1109/icra57147.2024.10611316

How many times has this paper been cited?

11 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?

Homayoun Honari, Mehran Ghafarian Tamizi, Homayoun Najjaran.

What is the DOI of this paper?

10.1109/icra57147.2024.10611316, permanently at https://doi.org/10.1109/icra57147.2024.10611316. 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.