Device-Edge Collaborative DNN Inference for Artificial Intelligence of Things

面向人工智能物联网的设备-边缘协同深度神经网络推理

摘要：Artificial Intelligence of Things (AIoT) has emerged as a new paradigm of deep integration between artificial intelligence (AI) and the Internet of Things (IoT) to support intelligent services. Benefiting from recent advancements in deep neural networks (DNNs), AIoT networks are expected to facilitate various applications, including smart agriculture, intelligent transportation, and smart cities. Despite the potential of DNN models, running them directly on AIoT devices presents challenges due to their limited computation, storage, and energy resources, which hinder the deployment of large-scale DNN models for high-precision task inference. To address the limitations of devices, device-edge collaborative DNN inference (CDI) has been envisioned as one of the forefront technologies for completing task inference. CDI involves segmenting a DNN model into two parts with initial layers processed by devices and subsequent layers handled by edge servers. Using this approach, the core idea of CDI is to transmit intermediate feature data (IFD) extracted from raw task data instead of transmitting the raw data of inference tasks. This workshop aims to establish a platform for disseminating cutting-edge research findings in device-edge CDI. Authors are welcome to submit original papers on topics including, but not limited to:

• Fundamental theory and performance analysis for device-edge CDI

• Network architectures for device-edge CDI

• Model partitioning and resource allocation for device-edge CDI

• Digital twin and metaverse for device-edge CDI

• Security and privacy issues in device-edge CDI

• Performance trade-off for device-edge CDI

• Device-edge CDI for drones and vehicular networks

• Experimental demonstrations and prototypes

一、会议信息

会议官网：https://www.academicenter.com/conferences/details/IEEE_ICCBD+AI2025.html
会议官邮：iccbdai@126.com
会议时间：2025年11月21日-23日
二轮截稿：2025年10月20日
会议地点：中国 · 厦门
主办单位：IEEE中国联合会、华侨大学
协办单位：华南理工大学、石河子大学、广州大学网络空间安全学院、中国网络空间新兴技术安全创新论坛（新安盟）、爱迩思出版社、ESBK国际学术中心、AC学术平台

Workshop 13专属投稿链接：https://ocs.academicenter.com/submission/stepone?conf_id=1894929943985827840&workshop_id=1963482102757879808

二、往届检索历史

三、征稿主题（集中但不限于“计算机、大数据与人工智能”等其他相关主题）

计算机：云计算、边缘计算、嵌入式计算、物联网、计算机体系结构和VLSI

大数据：大数据模型、处理算法及编程技术、多媒体大数据的表示、大数据分析 与测量、大数据持久化与保存、大数据的质量与源头控制、大数据保护、完整性 与隐私、大数据存储、大数据搜索与挖掘、大数据可视化分析、大数据和物联网、大数据开放平台、领域和行业的大数据

人工智能：神经网络和学习系统、计算机视觉、机器人及相关领域、知识工程、粗糙集理论、基础与应用、自然语言处理、模糊计算、进化计算、人工智能和物联网、建模和仿真中的人工智能

 

四、联系方式

会议官邮： iccvit@163.com 

 

会议咨询：+86 166 7013 6527 王老师（Tel/WeChat） 

 

邮箱咨询：ivy2895941090@foxmail.com