IEEE MSN 2024 Workshop on Distributed Machine Learning and Unlearning for Sensor-Cloud Systems (DLS2)

Workshop Organizers:

TPC Chairs:

Prof. Pengfei Wang
School of Computer Science and Technology, Dalian University of Technology
Email: wangpf@dlut.edu.cn

Prof. Fuliang Li
School of Computer Science and Engineering, Northeastern University
Email: lifuliang207@126.com

Prof. Geng Sun
College of Computer Science and Technology, Jilin University
Email: sungeng@jlu.edu.cn

Short Biography of Organizers:

Pengfei Wang (Member, IEEE) received the B.S., M.S. and Ph.D. degrees in software engineering from Northeastern University, China, in 2013, 2015 and 2020, respectively. From 2016 to 2018, He was a visiting Ph.D. with the Department of Electrical Engineering and Computer Science, Northwestern University, IL, USA. He is currently an Associate Professor with the School of Computer Science and Technology, Dalian University of Technology (DUT), China. He has authored more than 60 papers on high-quality journals and conferences, such as IEEE/ACM TON、IEEE TMC、IEEE JSAC、IEEE INFOCOM, ACM TOSN, IEEE TITS, and IEEE ComMag, etc. He also holds a series of patents in US and China. His research interests are in the area of Internet.

 

Geng Sun (Member, IEEE) received the B.S. degree in communication engineering from Dalian Polytechnic University, and the Ph.D. degree in computer science and technology from Jilin University, in 2011 and 2018, respectively. He was a Visiting Researcher with the School of Electrical and Computer Engineering, Georgia Institute of Technology, USA. He is an Associate Professor in College of Computer Science and Technology at Jilin University, and His research interests include wireless networks, UAV communications, collaborative beamforming and optimizations.

 

A Brief Description of Workshop:

        Sensor-cloud systems have become increasingly prevalent due to their capability to collect, store, and process vast amounts of data from various sensors. These systems leverage the power of cloud computing to handle substantial data influx, providing robust platforms for data analytics, machine learning (ML), and real-time decision-making. The integration of distributed ML techniques presents unique opportunities and challenges, particularly in terms of scalability, privacy, and real-time processing. Distributed ML can significantly improve the scalability of sensor-cloud systems, enabling enhanced performance and fault tolerance by distributing computational load across multiple servers. Additionally, privacy-preserving techniques like federated learning maintain data security by training models without sharing raw data, crucial in domains such as healthcare and finance.

        Moreover, the emerging concept of "unlearning" in ML, which focuses on the ability to remove specific data points from trained models, is gaining traction for its implications in privacy and data compliance. Unlearning supports compliance with data protection regulations by providing mechanisms to delete specific data points from models upon request, ensuring enhanced user privacy and trust. It also contributes to efficient data management by allowing selective data deletion without needing to retrain models from scratch. The Workshop on Distributed Machine Learning and Unlearning in Sensor-Cloud Systems aims to address these opportunities and challenges, bringing together researchers, practitioners, and enthusiasts to share their latest findings and ideas, fostering a collaborative environment to advance the state-of-the-art in this field.

        We invite high-quality submissions that address novel research, practical applications, and theoretical advancements in the following (but not limited to) topics:

          • Distributed machine learning algorithms and architectures for sensor-cloud systems

          • Scalability and performance optimization in distributed ML

          • Privacy-preserving techniques in distributed ML

          • Federated learning and its applications in sensor-cloud environments

          • Real-time data processing and analytics in sensor-cloud systems

          • Unlearning techniques in machine learning and their applications

          • Data compliance and regulation issues in ML and sensor-cloud systems

          • Case studies and applications of distributed ML and unlearning in various domains (e.g., healthcare, smart cities, environmental monitoring)

Planned Format of the workshop:

          The workshop will be presented in the forms of technical sessions.

        Duration of the workshop: Half-Day

        Tentative Schedule:
        09:00-10:00 am: Distributed Machine Learning
        10:10-11:00 am: Distributed Machine Unlearning

        Number of referred papers: 6-10

A Preliminary Call for Papers:

        Sensor-cloud systems have become increasingly prevalent due to their capability to collect, store, and process vast amounts of data from various sensors. These systems leverage the power of cloud computing to handle substantial data influx, providing robust platforms for data analytics, machine learning (ML), and real-time decision-making. The integration of distributed ML techniques presents unique opportunities and challenges, particularly in terms of scalability, privacy, and real-time processing. Distributed ML can significantly improve the scalability of sensor-cloud systems, enabling enhanced performance and fault tolerance by distributing computational load across multiple servers. Additionally, privacy-preserving techniques like federated learning maintain data security by training models without sharing raw data, crucial in domains such as healthcare and finance.

        Moreover, the emerging concept of "unlearning" in ML, which focuses on the ability to remove specific data points from trained models, is gaining traction for its implications in privacy and data compliance. Unlearning supports compliance with data protection regulations by providing mechanisms to delete specific data points from models upon request, ensuring enhanced user privacy and trust. It also contributes to efficient data management by allowing selective data deletion without needing to retrain models from scratch. The Workshop on Distributed Machine Learning and Unlearning in Sensor-Cloud Systems aims to address these opportunities and challenges, bringing together researchers, practitioners, and enthusiasts to share their latest findings and ideas, fostering a collaborative environment to advance the state-of-the-art in this field.

        We invite high-quality submissions that address novel research, practical applications, and theoretical advancements in the following (but not limited to) topics:

          • Distributed machine learning algorithms and architectures for sensor-cloud systems

          • Scalability and performance optimization in distributed ML

          • Privacy-preserving techniques in distributed ML

          • Federated learning and its applications in sensor-cloud environments

          • Real-time data processing and analytics in sensor-cloud systems

          • Unlearning techniques in machine learning and their applications

          • Data compliance and regulation issues in ML and sensor-cloud systems

          • Case studies and applications of distributed ML and unlearning in various domains (e.g., healthcare, smart cities, environmental monitoring)

Paper Submission Guidelines:

        Papers submitted to the workshop should be written in English conforming to the IEEE 2-column US-letter style IEEE Conference Template and submitted in pdf format. The paper should be submitted through EasyChair. Prospective authors are invited to submit full papers up to 6 pages (Long Papers, including tables, figure and references) in length. Accepted and presented papers will be included into the IEEE explore. Authors of accepted papers, or at least one of them, are requested to register and present their work at the conference, otherwise their papers will be removed from the digital libraries of IEEE after the conference.

Important dates:

• Sep. 20, 2024

  Paper Submission Deadline

• Oct. 20, 2024

  Paper Acceptance Notification

• Oct. 30, 2024

  Camera-Ready Paper submission Deadline

• Dec. 20-22, 2024

  Conference Date