AIS4CC

Description

The growing use of intelligent systems in dynamic and decentralized environments requires artificial intelligence (AI) services that can adapt to distributed computing infrastructures. The Cloud Continuum, born from the integration of edge and cloud computing, is a paradigm that seamlessly interconnects and combines the computing power and storage capacity of the cloud with the data processing capabilities of edge devices, IoT devices, and sensors. It enables the intelligent and context-aware distribution of tasks, supporting faster response times, lower latency, and more efficient use of network bandwidth.

In this model, devices are no longer just simple terminals but act as active nodes that can process data locally and contribute computational resources. This is particularly important in environments with intermittent connectivity or strict latency requirements. In particular, IoT devices and embedded sensors are key enablers of this architecture, generating continuous streams of data from environments such as smart cities, healthcare systems, transportation networks, and industrial facilities. As these edge and near-edge devices must often need to make autonomous decisions based on local data, there is a clear need for AI models that are lightweight, energy-efficient, and privacy-aware.

This workshop aims to bring together researchers and practitioners to discuss how AI services can be designed, distributed, and coordinated within the cloud continuum. As an example, a possible focus can be on developing reduction strategies and models that can operate autonomously at the edge, rather than constantly relying on the cloud. Independence is a primary need in situations such as autonomous vehicles, industrial control, and healthcare monitoring, where low latency and local decisions are necessary. Edge inference imposes important constraints, such as limited memory, limited processing resources, and compliance with data sovereignty and privacy regulations. The workshop will also focus on lightweight and energy-efficient models that can operate within these constraints while maintaining effectiveness, including strategies for dynamically transferring intelligence between cloud and edge (dynamic AI offloading) depending on available resources, operating conditions, and the urgency of processing. Other examples are federated learning techniques, which enable networked model training on decentralized data sources. Each node contributes to the global model by processing updates locally instead of forwarding sensitive data.

This workshop will explore all of the above issues and similar ones in order to enable AI services on the cloud continuum, enabling continuous adaptation, customisation, and resilience, allowing systems to operate autonomously even when connectivity is limited. 

We invite papers on, but not limited to, the following topics:

• Application of Federated Learning with AI-based services

• Compression methods (e.g., pruning, quantization) for deep learning models on the Cloud Continuum

• Lightweight AI-based services for monitoring in smart environments (healthcare organizations, industries, farming, cities,...)

• Lightweight deep learning models as services

• Generative AI on the cloud continuum

• LLMs on the cloud continuum

• Agentic AI coordination on the cloud continuum

• Privacy and ethical considerations in the implementation and deployment of AI services on edge devices (i.e., IoT sensors)

Important Dates

• Workshop Papers Submission: 30 September 2025

• Authors Notification: 24 October 2025

• Camera-ready Submission: 07 November 2025

• Workshop Date: 01 December 2025

All deadlines are in Anywhere on Earth time (AOE = GMT – 12).

Check the time in the AOE Zone here: https://time.is/Anywhere_on_Earth.

Submission and Publication Guidelines

Authors are invited to submit their papers (typically between 8 and 12 pages, including references and appendices) using Springer's LNCS (Springer LNCS format). Authors must upload their paper as PDF file using the easychair system by selecting "The 1st Workshop on AI Services,in the Computing Continuum" track.

Each paper will be reviewed by at least three members of the program committee to ensure high quality. Paper acceptance will be based on originality, significance, technical soundness, and clarity of presentation. All accepted papers will be included in the workshop proceedings published as part of the Lecture Notes in Computer Science (LNCS) series of Springer. A selection of accepted workshop papers will be invited to submit extended versions of their work for a potential publication in a special issue in the International Journal of Computer and Information System (IJCIS).

At least one author of an accepted paper must register and participate in the workshop. Registration is subject to the terms and conditions, procedure of the main ICSOC conference to be found at https://icsoc2025.hit.edu.cn.

Organizers

• Maria Fazio, Università degli Studi di Messina

• Massimo Mecella, Sapienza Università di Roma

• Cyril P. Wyon-Boyault, OneTreck

• Radu Prodan, Leopold Franzens-Universität Innsbruck

Publicity and Web Co-Chairs

• Flavia Monti, Sapienza Università di Roma

• Adriano Puglisi, Sapienza Università di Roma

Program Committee

TBC