Speakers: Dr Nuo Lei, Department of Electrical and Computer Engineering, National University of Singapore (NUS)
Date & Time: 22 October 2026, Thursday, 7:30-8:30pm
Venue: LT422, SDE3, National University of Singapore (NUS)
Abstract: Electric vehicles (EVs) exhibit significant nonlinearity, strong coupling, and multiple constraints, and their overall performance depends largely on how their powertrain designs and control strategies are optimized. In practical vehicle development, conflicting objectives are often considered simultaneously, resulting in complex multi-objective optimization problems. Evolutionary computation offers robust global search capabilities, while the resulting Pareto-optimal solution sets can provide diverse design options and decision support for different design preferences, thereby offering an effective framework for addressing complex optimization problems in EVs. This lecture first introduces the research background, fundamental principles, and main advantages of evolutionary computation in the field of EVs. It then systematically discusses its applications in powertrain component sizing and control strategy optimization using practical vehicle-development case studies, covering key stages such as problem formulation, optimization algorithm design, Pareto-front analysis, and simulation and experimental validation. The lecture further introduces the integration of evolutionary computation with data-driven approaches, including machine learning, reinforcement learning, and surrogate modeling, and explores its potential for achieving efficient, adaptive, and intelligent optimization in complex dynamic environments. Finally, the lecture examines challenges and future directions related to computational efficiency, cross-scenario transferability, and deeper integration with large language models and other advanced technologies, and summarizes the potential of evolutionary computation to advance transportation electrification.
*No registration required.
Speakers: Prof Cheung Yiu-Ming, Chair Professor (Artificial Intelligence), Department of Computer Science, Hong Kong Baptist University (HKBU)
Date & Time: 23 October 2026, Friday, 10:30-11:30am
Venue: 5-1 meeting room, SCIS 1, Singapore Management University (SMU)
Abstract: Although deep learning has made great progress, a good model often requires a large amount of artificially balanced and annotated data. Unfortunately, real-world data are often unbalanced, typically exhibiting a long-tailed distribution, which refers to a small number of classes with abundant training samples but the remaining large number of classes only with very few training instances. Under the circumstances, the performance of deep learning models trained on long-tailed data declines sharply in the tail classes. However, tail classes cannot be ignored in various situations such as rare disease diagnosis, and anomaly detection. Subsequently, long-tailed data is still very challenging to deep learning. In this talk, the impact of long-tailed data on deep learning models will be first introduced. Then, the research progress in this area will be reviewed, including some representative methods in the literature. Lastly, the potential research directions in this field will be discussed.
*No registration required.
The 2026 Shenzhen International FinTech Competition launches its first-ever International Track, open to full-time students enrolled at universities outside China (i.e., outside Mainland China, Hong Kong SAR, Macao SAR, and Chinese Taipei).
Registration has opened since 31 August. Submission deadline is 20 October 2026.
Shortlisted teams will be invited to Shenzhen for the Finals during 20-23 November 2026, with expenses covered.
See more details in the dedicated webpage here.
We will share the detailed information about the upcoming conferences once confirmed.