Specific Topics

We invite submissions of contributions, including insightful experiences, reviews, analyses, and new methods, related to deploying reinforcement learning (RL) in industrial, societal, and/or critical systems. Submissions may address, but are not limited to, the following key questions:

  • When should RL be used for such systems, and why? What problem characteristics make RL more suitable than other approaches? Considerations may include problem assumptions, mathematical properties, and deployment requirements.
  • How should RL agent performance and robustness be evaluated for real-world deployment? What methodologies should be used to compare RL with other approaches?
  • How can compliance with application requirements be ensured? What are the key challenges and trade-offs in terms of performance, complexity, and robustness?
  • How can we develop a common language to communicate effectively with decision-makers and domain experts?
  • What other practical challenges should practitioners consider when implementing RL in real systems?

Example Domains

Examples of industrial, societal, and critical systems include, but are not limited to:

  • Industrial processes
  • Transportation systems
  • Electricity and energy
  • Logistics
  • Healthcare
  • Finance
  • News recommendations
  • …

What Is Insightful Experience?

Insightful experiences include observations, lessons learned, and best practices derived from deploying—or attempting to deploy—an RL agent in a real system. These practical insights, gathered from both successes and failures, should contribute to answering the workshop’s main questions. Accepted papers will be presented during poster sessions. The authors of exceptional submissions will be invited to take part in the panel.

Submission Site

You can access the submission portal here.

Specifications and Template

We accept contributions ranging from 2 to 8 pages (excluding references) to accommodate different types of work. Please follow the RLC submission instructions available at here. The cover page is not mandatory.

Important Dates

  • Submission Deadline: May 30, 2025 (AoE, UTC-12) June 02, 2025 (AoE, UTC-12)
  • Author Notification: June 13, 2025 (AoE, UTC-12)
  • Camera-ready Deadline: June 27, 2025 (AoE, UTC-12)

Submission Policy

Double-Blind Reviewing

The reviewing process will be double-blind, meaning authors must anonymize their submissions. Author names, affiliations, and acknowledgments should not be included. While identifying information should generally be avoided, we acknowledge that anonymization may be challenging for submissions reporting insights and experiences. In such cases, we will apply flexibility when justified by the relevance and value of the contribution.

Dual-Submission Policy

Submissions to this workshop are non-archival, meaning we welcome ongoing and unpublished work, including papers under review at other venues. However:

  • Submissions focused on analysis, reviews, or new methods applied to well-established benchmarks must not have been previously published or accepted at any venue.
  • Submissions reporting insightful experiences or new applications are welcome even if previously published in 2024 or 2025.