15th Int IEEE (Tech Co-sponsor) Conf on Software, Knowledge, Information Management & Applications
(With the International Workshop-Cum-Training on Safety and Assurance of AI Systems)8-10 December 2023, Corus Hotel Kuala Lumpur, Malaysia (http://skimanetwork.org)
Prof. Victor Chang
Title: An Improved Federated Learning Algorithm for Privacy Preserving in Cybertwin-Driven 6G System
Prof. Victor Chang is a Professor of Business Analytics at Operations and Information Management, Aston Business School, Aston University, UK, since mid-May 2022. He was previously a Professor of Data Science and Information Systems at the School of Computing, Engineering and Digital Technologies, Teesside University, UK, between September 2019 and mid-May 2022. He has deep knowledge and extensive experience in AI-oriented Data Science and has significant contributions in multiple disciplines. Within 4 years, Prof Chang completed Ph.D. (CS, Southampton) and PGCert (Higher Education, Fellow, Greenwich) while working for several projects simultaneously. Before becoming an academic, he has achieved 97% on average in 27 IT certifications. He won 2001 full Scholarship, a European Award on Cloud Migration in 2011, IEEE Outstanding Service Award in 2015, best papers in 2012, 2015 and 2018, the 2016 European award: Best Project in Research, 2016-2018 SEID Excellent Scholar, Suzhou, China, Outstanding Young Scientist award in 2017, 2017 special award on Data Science, 2017-2023 INSTICC Service Awards, Talent Award Suzhou 2019, Top 2% Scientist 2017/2018, 2019/2020 & 2020/2021, the most productive AI-based Data Analytics Scientist between 2010 and 2019, Highly Cited Researcher 2021 and numerous awards mainly since 2011. He is ranked number 2 in distributed computing and number 42 in AI globally based on top 2% Scientists 2020 from Stanford University. Prof Chang was involved in different projects worth more than £14 million in Europe and Asia. He has published 3 books as sole authors and the editor of 2 books on Cloud Computing and related technologies. He published 1 book on web development, 1 book on mobile app and 1 book on Neo4j. He gave 35 keynotes at international conferences. He is widely regarded as one of the most active and influential young scientist and expert in IoT/Data Science/Cloud/security/AI/IS, as he has the experience to develop 10 different services for multiple disciplines. He is the founding conference chair for IoTBDS, COMPLEXIS, FEMIB and IIoTBDSC to build up and foster active research communities globally with positive impacts..
This keynote will present the latest research outputs for an Improved Federated Learning Algorithm for Privacy Preserving in Cybertwin-Driven 6G System. With the expected explosive use of the Internet of Everything in sixth generation (6G), the cybertwin network is able to convert user information to digital assets and provide extensive services. However, protecting and enhancing privacy of the processed and transmitted data in cybertwin-driven 6G is still in its infancy. Federated learning (FL) is a nascent distributed machine learning paradigm that is able to facilitate privacy protection in cybertwin networks. In a cybertwin network, imbalanced data distribution of the clients can increase the bias of the global model and sacrifice the performance of the FL model. Prior research work dealing with imbalanced data requires extra data information exchanged between clients and the server, which increases the risk of privacy leakage. To avoid privacy leakage, we design an estimation algorithm to determine the distribution of local data collected at the clients without the awareness of specific raw data. We consider two scenarios in FL: 1) the server could receive the individual trained model for each selected device and 2) the server could receive the aggregated model from the selected clients. We formulate two device selection problems to improve the training performance of the aforementioned scenarios. We develop two online learning algorithms to tackle the selection problems for both individual model uploading and aggregated model uploading. The proposed algorithms are conducted on the server, thereby avoiding privacy leakage and extra computation at the clients. We validate the effectiveness of the proposed client selection algorithms with sufficient experiments in cybertwin-driven 6G networks.
- Special session and tutorial proposal :
15 September 2023
- Full Paper Submission Deadline :
30 September 2023
- Notification of Paper Acceptance :
30 October 2023
- Camera Ready Paper Deadline :
10 November 2023
- Conference :
8-10 December 2023