杨旸,教授,IEEE Fellow,现任香港科技大学(广州)协理副校长(教学)、教育科学学院院长、物联网学域教授、港科广-特斯联“数字世界”联合研究中心主任。他的研究领域包括5G/6G移动通信系统、智能物联网、多层次算力网络、开放无线测试验证平台等,已申请了120多项科技发明专利,发表了300多篇学术论文,出版了六部中英文专著。杨旸牵头承担了国家重点研发计划“物联网与智慧城市关键技术及示范”重点专项项目《面向大湾区智慧城市群的5G泛在物联基础设施建设及示范》。
报告题目:协作边缘计算助力无线网络部署大型AI模型
报告摘要:Large AI models have emerged as a crucial element in various intelligent applications at the network edge, such as voice assistants in smart homes and autonomous robotics in smart factories. Computing big AI models, e.g., for personalized fine-tuning and continual serving, poses significant challenges to edge devices due to the inherent conflict between limited computing resources and intensive workload associated with training. Despite the constraints of on-device training, traditional approaches usually resort to aggregating data and sending it to a remote cloud for centralized computation. Nevertheless, this approach is neither sustainable, which strains long-range backhaul transmission and energy-consuming datacenters, nor safely private, which shares users’ raw data with remote infrastructures. To address these challenges, we alternatively observe that prevalent edge environments usually contain a diverse collection of trusted edge devices with untapped idle resources, which can be leveraged for edge training acceleration. Motivated by this, iwn this talk, we propose to leverage edge collaboration, a novel mechanism that orchestrates a group of trusted edge devices as a resource pool, for expedited, sustainable large AI model computing at the edge. As an initial step, we present a comprehensive framework for building collaborative edge computing systems and analyze in-depth its merits and sustainable scheduling choices following its workflow. To further investigate the impact of its parallelism design, we empirically study a case of four typical parallelisms from the perspective of energy demand with realistic testbeds. Finally, we discuss open challenges for sustainable edge collaboration to point to future directions of edge-centric large AI model computing.