Program

Tentative program


9.30 - 9.35	Opening
9.35 - 10.40	Keynote: Shahram Rahimi (University of Alabama), “Towards an Autonomous Patient Recommendation System”
10.40 - 11.00	Break
11.00 - 11.25	Adapting the Behavior of Reinforcement Learning Agents to Changing Action Spaces and Reward Functions. Raúl De la Rosa, Ivana Dusparic and Nicolás Cardozo
11.25 - 11.50	Lightweight Temporal Consistency for Grid-Based Obstacle Detection in Edge Devices. Omer Kurkutlu and Arman Roohi
11.50 - 12.15	Causal Knowledge Transfer for Multi-Agent Reinforcement Learning in Dynamic Environments. Kathrin Korte, Christian Medeiros Adriano, Sona Ghahremani and Holger Giese
12.15 - 12.30	Final discussion & Closing

Keynote

Towards an Autonomous Patient Recommendation System

Speaker: Shahram Rahimi, University of Alabama

Abstract. This work advances the development of an autonomous patient recommendation system by leveraging knowledge graphs (KGs) to map and analyze patient journeys. We introduce the Patient Journey Ontology (PJO) to systematically represent diagnoses, treatments, and outcomes, enabling the construction of interoperable Patient Journey Knowledge Graphs (PJKGs). Using large language models, clinical dialogues are automatically transformed into structured PJKGs that capture the complete trajectory of patient care. To power recommendations, we propose the Dynamic Feature and Temporal Similarity (DFTS) framework, which integrates feature based and temporal similarity with dynamic weighting, designed to work effectively even with limited healthcare data. A case study in chronic disease management demonstrates the system’s ability to identify comparable patient journeys and generate personalized recommendations. This work establishes a foundation for autonomous, data-driven decision support that enhances patient-centered healthcare delivery.

About the speaker:

Dr. Shahram Rahimi is the Department Head of Computer Science at The University of Alabama and serves as Chief Advisor to the Alabama Cyber Institute. He previously held the position of Gloria & Douglas Marchant Endowed Chair Professor and Head of the Department of Computer Science and Engineering at Mississippi State University, and prior to that was Professor and Chair of the Department of Computer Science at Southern Illinois University. Dr. Rahimi has a distinguished background spanning both academia and industry, including roles with NASA and IBM. A recognized expert in artificial intelligence, he has published more than 350 peer-reviewed articles and holds multiple patents and pending patents. He has served as Editor-in-Chief for major AI journals and sits on the editorial boards of several others. Dr. Rahimi has contributed to numerous federal task forces, advising on predictive analytics for policy and defense. His hybrid game-theory system, Foresight, is widely used as a decision-support tool in foreign policy. He also led the development of EmTime/Symphony, an intelligent algorithm for ER staffing now deployed in more than 1,000 emergency departments nationwide, and recognized by HealthTech magazine in 2018 as one of the top 10 AI healthcare technologies. Over the past two decades, Dr. Rahimi has organized 15 international conferences on computational intelligence and multi-agent systems and has served as principal investigator on more than $25 million in federally and industry-funded research projects.

Call for Papers

Modern computing systems are characterized by increasing heterogeneity and operate on larger and larger scales. Their complexity is hardly manageable by a human being, especially when it comes to taking timely decisions in highly dynamic environments or to guarantee strict Quality-of-Service requirements. With the rapid evolution of AI and ML techniques - including generative AI, agentic AI, and edge intelligence - new opportunities have emerged for designing more robust, sustainable, and secure computing systems. AI and ML techniques are increasingly adopted to assist or guide system self-adaptation, as they are used, e.g., to extract relevant information from highly dimensional and noisy monitoring data, to predict internal or external dynamics, to automatically plan (and possibly activate) adaptation actions.

However, there are still several challenges to face for researchers and practitioners aiming to take advantage of these methodologies and incorporate them in their systems. Fundamental issues towards the applicability of AI and ML techniques across diverse domains must be investigated, especially as regards the accuracy, robustness, explainability, safety, security, performance and sustainability of AI-driven autonomous computing systems.

In this workshop, we solicit high quality contributions that fit with the overarching theme of AI and ML meeting autonomous computing systems. We invite submissions of original research papers, as well as vision papers and experience reports.

Topics

The aim of the workshop is to share new findings, exchange ideas and discuss research challenges on the following topics (not an exhaustive list):

AI and ML techniques for self-* computing systems
Architectures and frameworks for AI integration
Sustainability aspects of AI-driven adaptation
AI ethics, bias mitigation, and trustworthiness in self-adaptive systems
Federated and multi-agent learning approaches for decentralized adaptation
Robustness, explainability, safety, and security of AI-driven computing systems
Integration of large language models (LLMs) and generative AI into autonomous computing systems
Edge intelligence and distributed decision-making in autonomous systems
Self-adaptation for AI/ML systems
Case studies and real-world implementations of AI for autonomous computing systems

Organizers

Valeria Cardellini

Tor Vergata University of Rome, Italy

Ilias Gerostathopoulos

Vrije Universiteit Amsterdam, The Netherlands

Stefano Iannucci

Roma Tre University, Italy

Gabriele Russo Russo

Tor Vergata University of Rome, Italy

Program Committee

Sherif Abdelwahed

Virginia Commonwealth University, USA

Jesse Ables

University of South Alabama, USA

Ivana Dusparic

Trinity College Dublin, Ireland

David Garlan

Carnegie Mellon University, USA

Sona Ghahremani

Hasso plattner Institute

Ann Gentile

Sandia National Laboratories, USA

Emilio Incerto

IMT School for Advanced Studies, Italy

Jialong Li

Tsinghua University, China

Matteo Nardelli

Bank of Italy

Raffaela Mirandola

Karlsruhe Institute of Technology, Germany

Sara Pederzoli

University of Modena and Reggio Emilia, Italy

Gregor Schiele

University of Duisburg-Essen, Germany

Author Information

All submissions are required to be formatted according to the standard IEEE Computer Society Press proceedings style guide. Papers can be submitted in PDF format via EasyChair, making sure to select the track “AI4AS-Workshop”. Submitted manuscripts must be no longer than 6 pages (including figures, tables, and references).

Accepted papers will be published in the ACSOS Companion volume and will appear in IEEE Xplore.

As per the standard IEEE policies, all submissions should be original, i.e., they should not have been previously published in any conference proceedings, book, or journal and should not currently be under review for another archival conference. We would like to also highlight IEEE’s policies regarding plagiarism and self-plagiarism, available here.

Moreover, as per IEEE guidelines, the use of content generated by artificial intelligence (AI) in a submission (including but not limited to text, figures, images, and code) shall be disclosed in the acknowledgments section.

3rd International Workshop on Artificial Intelligence for Autonomous computing Systems (AI4AS 2025)

Important Dates