AI on Edge: Training & Inference – Optimising AI Models Using Local Data on Resource-Constrained Edge Devices
Edge AI enables intelligent solutions deployed on edge devices, reducing latency, allowing offline execution, and providing strong privacy guarantees. However, bringing computational capabilities closer to data sources poses significant challenges in achieving both efficiency and accuracy for resource-constrained edge devices.
This workshop aims to explore the intricate trade-offs involved in optimising AI applications for edge environments. It addresses the critical need to balance high model accuracy with minimal resource consumption across diverse applications of Edge AI, including Healthcare, Autonomous Vehicles, Earth observation, Smart home and industrial automation, smart cities, Industrial Internet of Things (IIoT), generative AI (Gen AI), and large language models (LLMs).
Workshop Chairs
Ubotica, Ireland
Ubotica, Ireland
Ubotica, Ireland
Ubotica, Ireland
Ubotica, Ireland
Submission
Topics
The main workshop session focuses on, but is not limited to:
- Learning Paradigms: Exploration of diverse ML paradigms enabling analytics across the Edge-to-Cloud Continuum.
- Healthcare: Real-time diagnostics and patient monitoring.
- Autonomous Vehicles: On-board decision-making and navigation systems.
- Earth Observation: Environmental monitoring and disaster response.
- Smart Home: Automated and efficient household management through intelligent devices.
- Industrial Automation: Enhanced operational efficiency and predictive maintenance using edge AI.
- Smart Cities: Traffic management and public safety solutions.
- Industrial Internet of Things (IIoT): Predictive maintenance and manufacturing optimisation.
- Generative AI (Gen AI): Real-time content creation and adaptation.
- Large Language Models (LLMs): Localised natural language processing for privacy and efficiency.
- Lightweight model architectures: Optimised algorithms for efficient performance on resource-constrained devices.
Important Dates
- Workshop Papers Submission: 26th September 2024
- Authors Notification: 24th October 2024
- Camera Ready and Registration: 31st October 2024
- Workshop Date: 3rd December 2024
Submission Guidelines
We invite submissions in the following categories:
- Research Papers: Full papers with a 12-page limit, presenting original and novel research work. Submissions may include research results, evaluations, ongoing research, novel ideas with preliminary findings, advancements in the state of the art, or practical implementation experiences.
- Position Papers: Papers with an 8-page limit, offering position statements on issues related to the workshop topics. These papers should raise key questions, present controversial viewpoints, or discuss challenges and ideas to stimulate engaging discussions.
- Industry Papers: Full papers with a 12-page limit, focusing on industry-related research and developments. This track is designed for industry professionals and academics to showcase novel ideas, attract funding, and expand collaborations. Submissions must demonstrate a strong industrial association, such as the extensive evaluation of research results in industrial use cases or the active participation of industry representatives as co-authors.
Formatting and Submission
- All submissions must be in English and formatted according to the Springer LNCS guidelines.
- The page limits include the abstract, figures, and references.
- Papers must be submitted electronically as PDF files via the Workshop Submission System by the specified deadlines.
- Final camera-ready submissions should adhere to Springer’s LNCS formatting instructions.
Review Process
- Each paper will be peer-reviewed by at least three members of the program committee to ensure high quality.
- Papers will be evaluated based on originality, significance, technical merit, and clarity of presentation.
Publication
- All accepted papers will be included in the workshop proceedings, published by Springer.
Participation Requirement
- At least one author of each accepted paper is required to register for the workshop and attend to present the paper. Registration will follow the terms and conditions of the main ICSOC conference, details of which are available on their official website.
Best Paper Awards
Award Categories:
Submissions will be evaluated for originality, technical quality, clarity, and relevance to Edge AI. We particularly value research that introduces novel applications, tackles challenges in Edge AI deployment, or presents breakthroughs in the field. Award recipients will receive a certificate of recognition.