Invited talk:Split Learning for 6G Edge Intelligence
We invited Asst. Prof. Xianhao Chen, the Department of Electrical and Electronic Engineering, the University of Hong Kong, to give a academic talk, " Split Learning for 6G Edge Intelligence", for students and falculties of Soutwest Jiaotong University (SWJTU), and IEEE members within Chengdu. This event is technically co-sponsored by IEEE COMSOC Chengdu chapter.
Date and Time
Location
Hosts
Registration
- Date: 06 Jun 2024
- Time: 02:00 AM UTC to 04:00 AM UTC
-
Add Event to Calendar
- Xipu campus
- Southwest Jiaotong University
- Chengdu, Sichuan
- China
- Building: Internatioinal Cooperation Center
- Room Number: 2202
- Contact Event Hosts
-
Associate Professor Gang Liu
Key Lab. of Information Coding & Transmission, Southwest Jiaotong University
- Co-sponsored by Key Lab. of Information Coding & Transmission, Southwest Jiaotong University
Speakers
Weiguo Zhang of Xidian University
Corrector design of the true random number generator
The corrector of the true random number generator is a post-processing function used to reduce or eliminate the statistical weakness of the physical random number generator to improve the randomness and statistical independence of the output sequence. How to design the optimization compromise between correction ability, processing efficiency and nonlinear properties is the core problem in the design of true random number generator. For this problem, we study the design of the corrector from the perspective of the password function, using isometric linear code to construct a large class of output, college positive order, high nonlinearity, high algebra degree of nonlinear corrector, for the first time the correction order greater than the elastic order, implements the calibration output dimension, correction order, nonlinear degree, and algebra parameters between the optimization compromise.
Biography:
Zhang Weiguo, professor and doctoral supervisor of Xidian University. He received his bachelor's degree and doctor's degree from Xidian University in 2001 and 2007, respectively. His main research interests are symmetry cryptography, Boolean function design, pseudorandom sequences and their application. Personal home page: https://web.xidian.edu.cn/wgzhang
Email:
Address:Chengdu, China
Xianhao Chen of the University of Hong Kong
Split Learning for 6G Edge Intelligence
The next-generation mobile network aims to natively support distributed intelligence, such as federated learning, across massive wireless edge devices. Unfortunately, in the era of large models, the deployment of federated learning faces significant obstacles due to the limited resources on edge devices. In this talk, I will briefly introduce split learning (SL) and elucidate how it overcomes resource limitations via device-server co-training, thereby transforming next-generation edge AI. Then, I will present our recent work on adaptive split federated learning (AdaptSFL) in resource-constrained edge networks. Specifically, our work first provides a unified convergence analysis of split federated learning (SFL) to quantify the impact of model splitting and client-side model aggregation on the learning performance, based on which the AdaptSFL framework is developed to adaptively control model splitting and client-side model aggregation to balance communication-computing latency and training convergence in SFL. Simulation results demonstrate the effectiveness of our approach in accelerating SFL under resource constraints. At last, I will conclude the talk by discussing open problems in SL at the wireless edge.
Biography:
Xianhao Chen is now an assistant professor at the Department of Electrical and Electronic Engineering, the University of Hong Kong, and the Principal Investigator (PI) of the Wireless Information & Intelligence (WILL) lab. I obtained my Ph.D. degree in electrical and computer engineering from the University of Florida in 2022, supervised by Prof. Yuguang Fang, and my B.Eng. degree from Southwest Jiaotong University, Chengdu, China, in 2017, supervised by Prof. Gang Liu. My research interests include wireless networks, edge computing, distributed learning, and network security.
Email:
Address:Hong Kong, China
Agenda
June 6,10:00~12:00， 2024.
10:00~10:05, Opening remarks and introduction to the invited speakers, Associate Prof. Gang Liu
10:05~11:00, Invited talk: Split Learning for 6G Edge Intelligence
Speaker: Xianhao Chen (陈先浩), Assistant Professor, Department of Electrical and Electronic Engineering, The University of Hong Kong
Email: xchen@eee.hku.hk
11:00~12:00, Q&A