AI Advances in Edge Computing

A special issue of Mathematics (ISSN 2227-7390). This special issue belongs to the section "E1: Mathematics and Computer Science".

Deadline for manuscript submissions: 30 November 2025 | Viewed by 1583

Special Issue Editor


E-Mail Website
Guest Editor
Department of Computer Science, San Francisco State University, San Francisco, CA 94132, USA
Interests: B5G/6G network; edge computing; cyber security in intelligent IoT networks

Special Issue Information

Dear Colleagues,

Nowadays, with the rise in popularity of large language models (LLMs) and generative AI, more intelligent applications and research are being implemented in daily life and production. One significant trend is edge devices’ model compression and deployment to enhance rapid response, flexible customization with the application environments, and more controllable local data protection. Accordingly, various studies on model deployment and compression and cross-domain integration have been proposed to address these issues.

The increasing offering of new AI methods affords significant advances for edge computing and IoT applications. Examples of innovations unlocking new possibilities for resource-constrained AI at the edge include the following: a novel structured pruning approach which can greatly reduce complex CNN resource requirements without sacrificing accuracy; multi-compression scale DNN inference acceleration (MCIA), which uses cloud-edge-end collaboration and deep reinforcement learning; an optimization problem and algorithm, proposed for hosting LLM-powered generative AI on edge devices; and, finally, a method for the multistage low-rank fine-tuning of super-transformers (MLFS), which enables the parameter-efficient supernet training of LLMs, allowing the production of smaller models for edge applications at a constant cost.

This Special Issue will focus on the latest theoretical and computational studies on deploying intelligent models on computationally limited edge devices, with an emphasis on model compression, optimization, and application expansion. The topics include, but are not limited to, the following:

  • Theoretical research on large-model compression;
  • Performance optimization of model compression;
  • Applications of generative AI on edge devices;
  • Collaborative research on distributed multi-agent systems;
  • Security studies on edge intelligence;
  • Communication and collaboration in distributed multi-agent systems.

Dr. Qun Wang
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • edge AI
  • GenAI security
  • sparsity pruning
  • network compression
  • convolutional neural networks
  • edge computing
  • large language models
  • mixed sparsity

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (1 paper)

Order results
Result details
Select all
Export citation of selected articles as:

Research

18 pages, 351 KiB  
Article
Approximation of Time-Frequency Shift Equivariant Maps by Neural Networks
by Dae Gwan Lee
Mathematics 2024, 12(23), 3704; https://doi.org/10.3390/math12233704 - 26 Nov 2024
Viewed by 931
Abstract
Based on finite-dimensional time-frequency analysis, we study the properties of time-frequency shift equivariant maps that are generally nonlinear. We first establish a one-to-one correspondence between Λ-equivariant maps and certain phase-homogeneous functions and also provide a reconstruction formula that expresses Λ-equivariant maps [...] Read more.
Based on finite-dimensional time-frequency analysis, we study the properties of time-frequency shift equivariant maps that are generally nonlinear. We first establish a one-to-one correspondence between Λ-equivariant maps and certain phase-homogeneous functions and also provide a reconstruction formula that expresses Λ-equivariant maps in terms of these phase-homogeneous functions, leading to a deeper understanding of the class of Λ-equivariant maps. Next, we consider the approximation of Λ-equivariant maps by neural networks. In the case where Λ is a cyclic subgroup of order N in ZN×ZN, we prove that every Λ-equivariant map can be approximated by a shallow neural network whose affine linear maps are simply linear combinations of time-frequency shifts by Λ. This aligns well with the proven suitability of convolutional neural networks (CNNs) in tasks requiring translation equivariance, particularly in image and signal processing applications. Full article
(This article belongs to the Special Issue AI Advances in Edge Computing)
Show Figures

Figure 1

Back to TopTop