Welcome to AIoT 2023!

Nowadays, the impressive proliferation of IoT devices (predicted to reach 30 billion by 2030), able to monitor several real-world processes and environments, is driving the development of extreme analytics for business decisions based on the vast amount of data collected by smart objects. Indeed, emerging wireless technologies, such as 5G and LPWAN, are enabling the possibility to easily and efficiently connect tiny devices, which are also equipped with heterogeneous computational capacity, varying from smartphones to micro-controllers, deployed over large geographical areas. 

In such a context, emerging learning mechanisms, such as distributed and federated learning, can be a promising alternative to traditional centralized analytics. 

The First International ACM MobiHoc Workshop on the Integration between Distributed Machine Learning and the Internet of Things (AIoT) is specifically meant to gather new ideas, contributions, and experiences on the integration of Distributed and Federated Machine Learning with long-range IoT systems. 

Keynote: Opportunities and challenges of AI in the Internet of Vehicles

Speaker:  Giacomo Morabito,  University of Catania, Italy

McKinsey estimates that connected cars create up to 25 GB of data per hour. Such data can be utilized to realize AI-based applications in several contexts like predictive maintenance, safety, and energy efficiency. Furthermore, by enabling the exchange of appropriate portions of such data between cars and between cars and the elements of the road infrastructure, several other AI applications can be introduced that further increase safety on the roads and have the potential to mitigate traffic in our cities. 

The above is the premise of the exploitation of AI solutions in the Internet of Vehicles (IoV) which has captured the increasing attention of the industrial and academic R&D community. 

However, cars are only a portion of a bigger picture. In fact, urban mobility is rapidly changing pushed by the presence of affordable, electric, personal vehicles such as scooters and e-bikes. In fact, it is expected that the number of such alternative vehicles circulating in our cities will increase while the number of private cars will decrease.

Such “new” types of vehicles have characteristics and needs different from those of cars and offer interesting application scenarios for machine learning.

The objective of this talk is to provide an overview of the opportunities of the application of machine learning in an IoV including such alternative vehicles and of the challenges to be faced for their practical realization.

Speaker Bio

 
Giacomo Morabito received the laurea degree in Electronics Engineering and the Ph.D. in Electrical, Electronic, Computer, and Telecommunications Engineering from the University of Catania (Italy) in 1996 and 1999, respectively. From 1999 to 2001 he was with the Broadband and Wireless Networking Laboratory of the Georgia Institute of Technology (USA) as research engineer. In 2002 he joined the University of Catania, where he is professor of telecommunications, currently.

Call for Papers

Distributed and federated learning are nowadays popular techniques, as they promise to minimize the amount of unnecessary data streamed for processing and to move decisions closer to the data sources thus enabling faster, ideally real-time analytics. Moreover, the usage of distributed and federated learning techniques reduces the security risks associated with moving data and sustains energy-efficient total execution. 

The integration between Distributed/Federated Learning mechanisms and the Internet of Things poses a series of whole new challenges, such as compression of models to be transmitted over unreliable channels, optimization of the network lifetime, management of the scarce computation, communication and storage resources, to name a few.

AIoT, the First International Workshop on the Integration between Distributed Machine Learning and the Internet of Things,  is specifically meant to gather new ideas, contributions, and experiences on the integration of Distributed and Federated Machine Learning with long-range IoT systems. The workshop solicits original papers dealing with the open challenges in the integration between Distributed/Federated Learning and IoT, including theoretical works and practical experiences over emulated and/or real testbeds. Contributions on the optimization of Machine and Deep Learning over embedded IoT devices are also welcome.

Topics include, but are not limited to:

• Efficient Machine Learning in the IoT

• Hardware for Machine Learning and Deep Learning in the IoT

• Network Layer technologies to support Machine Learning in the IoT

• Protocols to support Distributed Machine Learning in the IoT

• Edge computing and IoT for distributed/federated learning

• Experimental validation of distributed machine learning for IoT

• Testbeds and tools for distributed machine learning in IoT

• Privacy-preserving data sharing and aggregation in distributed/federated learning

• Datasets and applications of distributed/federated learning in IoT (eg. spectrum sensing, healthcare, smart cities, and transportation)

• Scalability and performance issues in IoT and distributed/federated learning

• Emerging trends, challenges, and future directions in IoT and distributed/federated learning

Important Dates

• Paper submission: July 12, 2023 August 5, 2023 (11:59pm US Eastern Time) 

• Acceptance notification: August 12, 2023 

• Camera ready and registration: August 30, 2023

• Workshop date: October 23, 2023

Submission Instructions

Papers should be submitted via the HotCRP submissing website (https://aiot23.hotcrp.com/).

Submissions must be original, unpublished work, and not currently under consideration elsewhere. Papers should not exceed 6 pages (US letter size) double column including figures, tables, and references in standard ACM format. Papers must be submitted electronically in printable PDF form. Templates for the standard ACM format can be found at this link: http://www.acm.org/publications/article-templates/proceedings-template.html . If you are using LaTeX, please refer to the sample file “sample-sigconf.tex” after you download the .zip templates file and unzip it. Note that the document class “\documentclass[sigconf]{acmart}” should be used. No changes to margins, spacing, or font sizes are allowed from those specified by the style files. Papers violating the formatting guidelines will be returned without review.

All submissions will be reviewed using a single-blind review process. The identity of referees will not be revealed to authors, but author can keep their names on the submitted papers, on figures, bibliography, etc.

Camera Ready and Registration Instructions

-Camera-Ready Preparation: Authors should visit the official ACM MobiHoc 2023 website at https://www.sigmobile.org/mobihoc/2023/camera-ready.html to follow the procedure for preparing the camera-ready version of their paper. If there are any formatting errors in your camera-ready, please upload a new version to address the formatting errors soon. 

-Author Registration Requirement: To ensure the inclusion of the paper in the final proceedings, every paper must have a separate regular (i.e., non-student) registration by an author. ONE registration is valid for ONE paper. If an author has both a conference paper and a workshop paper, they can use the combined conference+workshop rate to cover both papers, but they must be authored by the same individual. Authors should register prior to the submission of camera-ready manuscript. The registration policy and site can be accessed at https://www.sigmobile.org/mobihoc/2023/registration.html.  

-Confirmation Email: After completing the registration process, authors should send an email to the Registration Chair, Laura Galluccio (laura.galluccio@unict.it). The subject of the email should be "[ACM Mobihoc 2023 Registration]." In the body of the email, authors need to provide the following information:  

• Paper number 

• Paper title 

• The name of the registered author  

This confirmation email is necessary to verify the name of the author who is registered for each paper.

-Paper Presentation: Each paper must be presented at the workshop. We recommend you plan your trip as soon as possible. Venue and Hotel information is available at https://www.sigmobile.org/mobihoc/2023/venue.html. Visa information is available at https://www.sigmobile.org/mobihoc/2023/visa.html

Dual Submission Policy

Accepted papers will appear in the conference proceedings published by the ACM. Warning: It is ACM policy not to allow double submissions, where the same paper is submitted to more than one conference/journal concurrently. Any double submissions detected will be immediately rejected from all conferences/journals involved.