Sian Lun Lau

Dr. Sian Lun Lau received his Dr.-Ing. and MSc in Electrical Communication Engineering from the University of Kassel, Germany. He also holds a BEng with Hons in Electronics and Telecommunications Engineering from Universiti Malaysia Sarawak (UNIMAS).

During his nine years (2004 – 2013) as a researcher at the Chair for Communication Technology (ComTec) at the University of Kassel, he has worked and managed various German National- and EU-funded research projects. Among them are EU IST-MobiLife, ITEA S4ALL, BMBF MATRIX and EU-SEAM4US.

He joined Sunway University, Malaysia, in February 2013 as a senior lecturer and Head of the Department of Computing and Information Systems until March 2021. He is currently a Professor at the Department of Smart Computing and Cyber Resilience at the Faculty of Engineering and Technology.

He is currently a senior member of the Institute of Electrical and Electronics Engineers (IEEE) and serves as the Vice Chair of the IEEE Computer Society Malaysia Chapter for the term 2023/2024 and 2025/2026. His research interests include ubiquitous computing, sustainable smart city, context-awareness, and applied machine learning. His recent research projects include EDUFI ImpactXChange, ISPF BEST, EDUFI SustHack, MOSTI TED2 RADIC and US DOD DeepSpray+.

Contact: sianlunl@sunway.edu.my

x/twitter: @wahlau

Facebook: https://www.facebook.com/dr.lausl

Linkedin: https://www.linkedin.com/in/sianlun/

Abstracts

Context-Aware Digital Twins for Smart and Sustainable Cities: Bridging Physical and Virtual Urban Environments

Abstract: Smart and sustainable cities require seamless integration between physical infrastructure and digital representations to enable intelligent decision-making and resource optimisation. This talk presents a comprehensive framework for developing context-aware digital twins that leverage pervasive computing technologies to create dynamic, real-time models of urban environments.

Drawing from research in context awareness and pervasive computing, we explore how smartphone-based activity recognition, IoT sensor networks, and crowd-sourced data feed into digital twin architectures. The presentation demonstrates practical applications including traffic flow optimisation, energy management, and citizen behaviour modelling for urban planning.

Key contributions include novel approaches for multi-modal sensor data fusion, maintaining synchronisation between physical and digital environments, and privacy-preserving data collection. Case studies from different smart city initiatives illustrate real-world deployment challenges and solutions for sustainable urban development.

Sustainability-Driven Innovation: Using Hackathons to Cultivate Environmental Mindset in Engineering and Technology Students

Abstract: Engineering education faces the critical challenge of preparing students to address global sustainability issues while fostering innovation capabilities. This talk presents empirical findings on how hackathons can effectively cultivate environmental awareness and sustainable thinking among engineering and computing students. Our research examines the SustHack methodology, which combines collaborative problem-solving with sustainability-focused challenges. Through systematic evaluation of student engagement, behavioural changes, and solution quality, we demonstrate how hackathons serve as powerful pedagogical tools for sustainability education.

Key findings include improved environmental consciousness, enhanced collaborative skills, and increased motivation to pursue green technology solutions. The presentation discusses implementation strategies, assessment frameworks, and scalability considerations for education institutions and beyond. This approach addresses the urgent need for sustainability-minded engineers capable of driving the region's transition toward sustainable development goals and green economy initiatives.

Sustainable AI for Industrial Applications: Energy-Efficient Computer Vision Through Small Sample Learning

Abstract: Industrial sustainability requires intelligent systems that minimise computational resources while maintaining performance. This talk presents sustainable AI approaches using computer vision with focus on energy-efficient model training and deployment.

Our research demonstrates how semi-supervised learning and self-learning techniques achieve robust performance with minimal labeled data, significantly reducing training energy consumption. Key innovations include pseudo-labelling methods, small sample learning strategies, and lightweight architectures.

The presentation emphasises sustainable AI principles: efficient data utilisation, reduced annotation requirements, and sustainable deployment.