Department of Applied Mathematics,
National Yang Ming Chiao Tung University, Taiwan
My Journey of Generative AI and TAIDE: Trustworthy AI Dialogue Engine
Since its launch on November 30, 2022, OpenAI's Chat-GPT has heralded a new era in AI that has captivated the world. The integration of AI into our lives has become a learning curve that we are all adapting to. In my presentation, I aim to highlight the pivotal advancements in Generative AI (GAI) from my perspective. Specifically, I will showcase a customer service assistant system that operates with a human-in-the-loop approach akin to RLHF (Reinforcement Learning with Human Feedback). Furthermore, the latter part of my talk will introduce the TAIDE (Trustworthy AI Dialogue Engine) project, an NSTC-granted initiative based in Taiwan. I will provide an update on the current status of TAIDE and invite your invaluable suggestions and feedback. This discussion aims to spotlight the advancements and aspirations of the TAIDE project, emphasizing the need for collaborative insights and inputs for its continual development.
Prof. Yuh-Jye Lee is a computer scientist who holds a PhD in Computer Sciences from UW-Madison (2001). He has held positions at esteemed universities, including National Chun Cheng University, Taiwan Tech, and National Yang Ming Chiao Tung University.
Prof. Lee also had the opportunity to serve the government from November 2020 to January 2023 as the Deputy Executive Secretary of the Office of Science and Technology Policy, National Science and Technology Council. This office is responsible for Taiwan's science and technology policies, budget allocation, and the review of national-level science and technology projects. Currently, he serves as the Chief Executive Officer of the Taiwan Information Security Center at the Research Center for Information Technology Innovation, Academia Sinica.
Prof. Lee's research primarily focuses on AI, data science, machine learning, federated learning, and information security, with a specific interest in developing lightweight anomaly detection techniques for IoT applications. Over the past decade, he has developed numerous efficient machine learning algorithms such as SSVM and RSVM that have been successfully applied in various fields, including breast cancer diagnosis and prognosis, network intrusion detection, and fake news prevention. In essence, his research is centered on AI in security and security in AI. He is also in charge of the TAIDE project, which stands for Trustworthy AI Dialogue Engine.