IoT Thrust Seminar | Advancing Frontier of Efficiency-Thoroughness Trade-off (ETTO) for AIoT Sensing
Supporting the below United Nations Sustainable Development Goals:支持以下聯合國可持續發展目標：支持以下联合国可持续发展目标：
The number of IoT applications is expected to grow continuously in a high speed. The applications cover a wide range of areas including manufacturing, utilities, transportation, healthcare, and automotive. With the advancements of low-cost edge hardware accelerators, data processing and decision making at the edge using artificial intelligence (AI) technologies for time sensitive applications are made possible. AI’s abilities to automatically extract useful features from complex physical processes based on big data have enhanced performance for many IoT sensing applications. However, the complexity of AI’s structure may also incur thoroughness considerations, such as generalizability, privacy, security, safety, domain shift, etc. This talk gives the essential ongoing challenges for AIoT sensing and the developments of advancing the pareto frontier for AIoT sensing.
Qun Song received B.Eng from Nankai University. After that, she joined the NTU IoT Sensing Group of Nanyang Technological University, Singapore as a PhD student supervised by Dr. Rui Tan. Her research focuses on developing resilient machine learning for Internet of Things and cyber-physical systems. She has published multiple papers on conferences such as ACM UbiComp, ACM SenSys, ACM MobiCom, ACM/IEEE IPSN, and ACM e-Energy, as well as on prestigious journals such as ACM Transactions on Sensor Networks, IEEE Transactions on Mobile Computing, and IEEE Transactions on Smart Grid.