In recent years, the widespread integration of machine learning and deep learning models into real-world applications across various domains has exposed numerous operational challenges in model building, deployment, monitoring, and maintenance.
Machine Learning Operations (MLOps) has emerged as a key approach to addressing these challenges. It encompasses practices and tools designed to streamline the entire ML lifecycle, with a focus on end-to-end automation, reproducibility, and scalability of ML workflows.
This workshop encourages discussions on bridging the gap between data science experimentation and the reliable operation of ML components in production. We invite submissions that address the open challenges, latest breakthroughs, and practical use cases in the rapidly evolving field of MLOps. By bringing together researchers and practitioners, the workshop aims to foster the exchange of valuable insights and advancements from both academia and industry.
Topics of Interest
Submissions are encouraged on, but not limited to, the following topics:
- MLOps frameworks
- ML systems lifecycle management
- ML pipelines orchestration
- Practices to ensure ML model reproducibility, traceability, and explainability
- Continuous integration/continuous delivery (CI/CD) practices for ML models
-
ML model monitoring and observability
- Application of MLOps principles to large language models (LLMOps)
- ML-specific architecture design and patterns
- Experience reports on real-world MLOps applications
- Challenges in applying MLOps to specific domains (e.g., healthcare and finance)
- Ethics and Accountability in MLOps
- AutoML applications in MLOps
-
Collaboration and team dynamics in MLOps
- Regulatory and policy aspects of MLOps
- MLOps strategies for Green AI
- Security and data privacy in MLOps
Important Dates (AoE)
- Paper submission deadline: Tuesday, May 20, 2025
- Notification of acceptance: Thursday, June 26, 2025
- Camera-ready submission: Tuesday, 26 August 2025
- Workshop date: TBD (during ECAI 2025)
Review Criteria
Submissions will be evaluated based on:
- Originality and novelty
- Research and industrial relevance
- Technical quality and soundness
- Clarity of presentation
Submission Guidelines
We invite two types of submissions:
- Research Papers: Original research contributions advancing the state-of-the-art in MLOps.
- Full-papers (up to 10 pages + 2 for references)
- Short-papers (up to 4 pages + 1 for references)
- Experience Reports (up to 6 pages + 1 for references): Reports on practical applications, lessons learned, and case studies related to MLOps implementation in real-world scenarios.
Submitted work must be original, previously unpublished, and not under consideration or review by any other publication venue. All submissions must be formatted according to the ECAI 2025 formatting guidelines and submitted via the workshop’s submission system (submission link forthcoming). Papers will undergo a double-blind review process (authors’ names must be omitted from the subission), with each manuscript receiving at least two evaluations from the program committee.
We look forward to your contributions and to the stimulating discussions on MLOps that will unfold at the workshop!