Elsevier Future Generation Computer Systems - Special Issue on On-device Artificial Intelligence solutions with applications on Smart Environments

Dear Colleagues,

we would like to invite you to contribute a paper to the Special Issue on =E2=
=80=9COn-device Artificial Intelligence solutions with applications on Smart=
 Environments=E2=80=9D, which is hosted by Elsevier Future Generation Comput=
er Systems (FGCS).


Motivation and Scope

The recent advancements in Artificial Intelligence (AI) and the increasing c=
omputational power acted as catalyzer for the widespread diffusion of Intell=
igent Cyber Physical Systems (ICPSs) as a novel way to run smart application=
s with a =E2=80=9Creasoning=E2=80=9D component. Unfortunately, the limited h=
ardware capabilities of these devices pose significant limitations on the co=
mplexity of the tasks and Deep Learning models that can be run effectively. =
During the years, solutions like weights compression or quantization have be=
en proposed to address this issue, but they usually require a careful tuning=
 and most of the time they consist in a post-training operation. From the ve=
ry beginning, the training of complex Deep Learning models has always been r=
eserved to powerful machines with large computing capabilities (typically id=
entified in the Cloud), limiting the Edge only to the inference.  However, t=
hese solutions do not work especially when latency, security, and high custo=
mization aspects become key prerequisites. In such a context emerges the nee=
d of novel methods to deliver the intelligence into an embedded system witho=
ut the data leaving the device. Originally born as a complementary technolog=
y, On-device AI is expected to become a hot topic in the next years as a new=
 paradigm where both training and inference processes are performed on the s=
ame device. If on the one hand, the possibility to run intelligent algorithm=
s on these systems is a challenging task, on the other the benefits in terms=
 of response time and energy efficiency derived from this technology are goi=
ng to be the foundations for a novel type of =E2=80=9Creasoning=E2=80=9D sys=
tems. To this aim novel architectures and frameworks should be explored to e=
nable the access to AI based tailored services. Considering a scenario where=
 the Edge would potentially store sensitive data (that should never travers =
the Internet), it is evident that these devices could become the target of a=
ttacks by malicious users. In this sense, privacy and security aspects repre=
sent another key elements to be carefully considered and implemented. This s=
pecial issue has the goal to promote original, unpublished, high-quality res=
earch about On-device AI solutions applied to the Smart Environments and Ind=
ustry 4.0 contexts.

The topics of interest include, but are not limited to:
=E2=80=A2 On-device training solutions
=E2=80=A2 On-device AI applications
=E2=80=A2 Federated Learning training and inference strategies on Edge devic=
es
=E2=80=A2 AI Intelligent Systems
=E2=80=A2 AI for Microcontrollers
=E2=80=A2 AI applications at the Edge
=E2=80=A2 AI methods for Industrial applications
=E2=80=A2 AI based services at the Edge
=E2=80=A2 Hardware efficient Deep Learning applications
=E2=80=A2 Energy efficient Deep Learning algorithms
=E2=80=A2 Privacy and Security for Deep Learning
=E2=80=A2 Comparative analysis of on-device AI frameworks
=E2=80=A2 Implementation case studies
=E2=80=A2 Low-power AI applications and methods
=E2=80=A2 Lightweight AI algorithms for Edge devices
=E2=80=A2 Edge architectures and frameworks for AI

Guest Editors

Fabrizio De Vita, University of Messina, [email protected]<mailto:fdevita@uni=
me.it>
Dario Bruneo, University of Messina, [email protected]<mailto:[email protected]=
t>
Sajal K. Das, Missouri University of Science and Technology, [email protected]<ma=
ilto:[email protected]>

Important Dates

=E2=80=A2 Submission portal opens: July 25, 2024
=E2=80=A2 Deadline for paper submission: February 25, 2025
=E2=80=A2 Latest acceptance deadline for all papers: February 25, 2025


Manuscript Submission Instructions

The FGCS=E2=80=99s submission system (Editorial Manager: https://www2.cloud.=
editorialmanager.com/fgcs/default2.aspx)  will be open for submissions to ou=
r Special Issue from July 25, 2024. When submitting your manuscript please s=
elect the article type VSI: On-device AI.

All submissions deemed suitable by the editors to be sent for peer review wi=
ll be reviewed by at least two independent reviewers. Once your manuscript i=
s accepted, it will go into production to be published in the special issue.=


Best regards,
The Guest Editors

    ------------------------------------------------------------------------=
----------------
    To unsubscribe from CHI-RESOURCES send an email to:
     mailto:[email protected]

    To manage your SIGCHI Mailing lists or read our polices see:
     https://sigchi.org/operations/listserv/
    ------------------------------------------------------------------------=
----------------