Conference paper · 2026 · Open access

Safe Reinforcement Learning with Contrastive Risk Prediction

Hanping Zhang, Yuhong Guo

Canadian Conference on Computer and Robot Vision · Published 14 May 2026 · DOI 10.21428/d82e957c.6bf94655

Open full textCite this paper

In brief

“Safe Reinforcement Learning with Contrastive Risk Prediction” is a 2026 conference paper by Hanping Zhang and Yuhong Guo, published in Canadian Conference on Computer and Robot Vision. OpenAlex counts 4 citations as of 22 November 2026. It is open access under a CC BY licence: the full text is free to read at Canadian Conference on Computer and Robot Vision.

Citations
4
Most cited in
2024 (3)
References
22
Authors
2

Abstract

As safety violations can lead to severe consequences in real-world applications, the increasing deployment of reinforcement learning (RL) in safety-critical domains such as robotics has propelled the study of safe reinforcement learning (safe RL). In this work, we propose a risk preventive training method for safe RL, which learns a binary classifier based on contrastive sampling to predict the probability of a state-action pair leading to unsafe states. Based on predicted risk probabilities, risk preventive trajectory exploration and reward shaping-based optimality criterion modification are simultaneously conducted to induce safe RL policies. We conduct experiments in robotic simulation environments. The results show the proposed approach outperforms existing model-free safe RL approaches, and yields comparable performance with the state-of-the-art model-based method.

Cite this paper

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

APA 7

Zhang, H., & Guo, Y. (2026). Safe Reinforcement Learning with Contrastive Risk Prediction. Canadian Conference on Computer and Robot Vision. https://doi.org/10.21428/d82e957c.6bf94655

MLA 9

Zhang, Hanping, and Yuhong Guo. “Safe Reinforcement Learning with Contrastive Risk Prediction.” Canadian Conference on Computer and Robot Vision, 2026. https://doi.org/10.21428/d82e957c.6bf94655.

Chicago

Zhang, Hanping and Yuhong Guo. 2026. “Safe Reinforcement Learning with Contrastive Risk Prediction.” Canadian Conference on Computer and Robot Vision. https://doi.org/10.21428/d82e957c.6bf94655.

Harvard

Zhang, H. and Guo, Y. (2026) ‘Safe Reinforcement Learning with Contrastive Risk Prediction’, Canadian Conference on Computer and Robot Vision. doi:10.21428/d82e957c.6bf94655.

IEEE

H. Zhang and Y. Guo, “Safe Reinforcement Learning with Contrastive Risk Prediction,” Canadian Conference on Computer and Robot Vision, 2026, doi: 10.21428/d82e957c.6bf94655.

BibTeX

@inproceedings{zhang2026safe,
  title     = {{Safe Reinforcement Learning with Contrastive Risk Prediction}},
  author    = {Zhang, Hanping and Guo, Yuhong},
  booktitle = {Canadian Conference on Computer and Robot Vision},
  year      = {2026},
  doi       = {10.21428/d82e957c.6bf94655},
  url       = {https://doi.org/10.21428/d82e957c.6bf94655}
}

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
  • risk prediction
  • binary classifiers
  • reward shaping
  • 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?

Zhang, H., & Guo, Y. (2026). Safe Reinforcement Learning with Contrastive Risk Prediction. Canadian Conference on Computer and Robot Vision. https://doi.org/10.21428/d82e957c.6bf94655

How many times has this paper been cited?

4 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 Canadian Conference on Computer and Robot Vision: sign in and open the full text from this page.

Who wrote this paper?

Hanping Zhang, Yuhong Guo.

What is the DOI of this paper?

10.21428/d82e957c.6bf94655, permanently at https://doi.org/10.21428/d82e957c.6bf94655. 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.