| Prof. Jiu LiuHarbin Engineering University, China Experience: Liu Jiu is a professor and master's supervisor in the Department of Law at the School of Humanities and Social Sciences, Harbin Engineering University. He holds a PhD in International Law from China University of Political Science and Law and a joint PhD from the University of California, Davis. His research interests include nuclear energy law and international law. Since July 2016, he has been employed at Harbin Engineering University, serving as a lecturer, associate professor, and later promoted to professor (with promotion synchronized with the completion of his project). He has led the National Social Science Fund Youth Project "Research on the Legislation of China's Nuclear Damage Compensation System" (2018-2023) and other projects such as the Heilongjiang Provincial Philosophy and Social Science Research Planning Project. In 2014, he went to study at the University of California, Davis (September 2014 - September 2015), and in 2019, he participated in the International Atomic Energy Agency (IAEA) Regional Directors Meeting for Africa (Marrakesh, Morocco). He has published a monograph titled "Research on the Deposit Insurance System: Taking the American Experience and International Guidelines as a Starting Point" and an article titled "On the Construction of China's Nuclear Damage Liability System in the Context of the Atomic Energy Law" in "Shanghai Law Research". In 2021, he participated in the International Nuclear Law Training organized by the OECD Nuclear Energy Agency, and in June 2025, he represented the university to participate in the IAEA Nuclear Law Seminar and conducted research at Hainan Nuclear Power. He also serves as a council member of the International Economic Law Research Association of the Beijing Law Society (since 2016) and a member of the Harbin Municipal Committee of the Chinese People's Political Consultative Conference (2022 session). Title: The Development of Nuclear Industry in China Aiming at Carbon Peaking and Neutrality: Value, Risks and Response of Law |
| Assoc. Prof. Aslina Binti BaharumExperience:Ts. Dr Aslina Baharum is an Associate Professor and UX Researcher at the School of Engineering and Technology, Sunway University. Previously, she was a Senior Lecturer at Universiti Teknologi MARA (UiTM), and Universiti Malaysia Sabah (UMS). She also has industry experiences where she worked as an IT Officer for the Forest Research Institute of Malaysia (FRIM). She had experienced more than 20 years in the IT field. She received a PhD in Visual Informatics (UKM), a Master Science degree in IT (UiTM) and graduated Bachelor of Science (Hons.) in E-Commerce from UMS. She is a member of the Young Scientists Network - Academy of Science Malaysia, Senior Member IEEE, and certified Professional Technologist from MBOT, and served as MBOT/MQA auditor. Title: When Algorithms Decide: Human Factors at the Heart of Digital Risk |
 | Assoc. Prof. Yi-Zeng Hsieh National Taiwan University of Science and Technology
Speech Title: From Forecasting to Autonomous Energy Orchestration: Transformer-Driven and Generative AI–Empowered Smart Power Dispatch for Intelligent Buildings Abstract: The accelerating impact of climate change and the sustained growth of electricity demand have made intelligent energy management in commercial buildings a critical challenge for modern smart cities. Office buildings, in particular, exhibit highly volatile short-term power consumption patterns due to complex interactions among environmental conditions, human activities, and the operation of large-scale equipment such as HVAC and chiller systems. In this keynote, we present a unified framework that bridges accurate time-series power forecasting and autonomous decision-making for power dispatch by integrating Transformer-based deep learning models with generative artificial intelligence. We introduce a lightweight yet effective forecasting architecture, termed GCT (Gated CNN-Transformer), which combines Gated Residual Networks for temporal feature selection, convolutional neural networks for local spectral pattern extraction, and a Transformer encoder for long-range dependency modeling and sequence reconstruction. The proposed model achieves superior prediction accuracy with significantly fewer parameters compared to conventional RNN and LSTM baselines, enabling practical deployment on edge computing platforms. Beyond prediction, this work demonstrates how generative AI can be elevated from a passive analysis tool to an active operational agent. By feeding short-term load forecasts into a large language model–based reasoning engine, the system can automatically generate interpretable and actionable power dispatch strategies under predefined operational constraints, effectively transforming building energy management from a reactive to a proactive and autonomous paradigm. Through extensive experiments on public benchmarks and real-world office building datasets, as well as a fully implemented web-based and Dockerized deployment, this talk illustrates a new vision for closed-loop, AI-driven energy management systems. The presented approach highlights how the convergence of Transformers and generative AI can reshape the future of intelligent buildings, digital twins, and sustainable energy orchestration in smart infrastructures. |
