Speaker: Yaoxin Wu, Assistant Professor, Eindhoven University of Technology, The Netherlands
Date: 8 July 2024
Time: 9:00 - 10:00am (Central European Summer Time, UTC+2) [Convert to your local time]
Zoom link: https://vuw.zoom.us/j/96385586267
Dr. Yaoxin Wu is an assistant professor in the Information Systems group at Eindhoven University of Technology. He received his Ph.D. degree in computer science from Nanyang Technological University, Singapore, in 2023. He was a Research Associate with the Singtel Cognitive and Artificial Intelligence Lab for Enterprises (SCALE@NTU). He has published research papers in leading journals (IEEE TNNLS, IEEE TKDE, IEEE TITS, IEEE TVT, etc.) and top-tier AI conferences (NeurIPS, ICML, ICLR, AAAI, UAI, AAMAS, etc.). In addition, he has served as PC member in NeurIPS, ICML, ICLR, AAAI, IJCAI, ECAI, etc, and Area Chair in IEEE Conference on Artificial Intelligence. He has also served as reviewer for prestigious journals in AI/OR fields, such as Transportation Research Part E, IEEE Transactions on Cybernetics, Annals of Operations Research, Transactions on Neural Networks and Learning Systems, etc. His research interests mainly include deep learning, combinatorial optimization, multi-objective optimization, and integer programming.
Existing deep reinforcement learning (DRL) methods for multi-objective vehicle routing problems (MOVRPs) typically decompose an MOVRP into subproblems with respective preferences and then train policies to solve corresponding subproblems. However, such a paradigm is still less effective in tackling the intricate interactions among subproblems, thus holding back the quality of the Pareto solutions. To counteract this limitation, we introduce a collaborative deep reinforcement learning method. We first propose a preference-based attention network (PAN) that allows the DRL agents to reason out solutions to subproblems in parallel, where a shared encoder learns the instance embedding and a decoder is tailored for each agent by preference intervention to construct respective solutions. Then, we design a collaborative active search (CAS) to further improve the solution quality, which updates only a part of the decoder parameters per instance during inference. In the CAS process, we also explicitly foster the interactions of neighboring DRL agents by imitation learning, empowering them to exchange insights of elite solutions to similar subproblems. Extensive results on random and benchmark instances verified the efficacy of PAN and CAS, which is particularly pronounced on the configurations (i.e., problem sizes or node distributions) beyond the training ones.
Speaker: Yuejiao Gong, South China University of Technology, China
Date: 12 August 2024
Time: 4:00 - 5:00pm (Beijing Time, UTC+8) [Convert to your local time]
Zoom link: https://vuw.zoom.us/j/96570642754
Yue-Jiao Gong is a Professor at the School of Computer Science and Engineering, South China University of Technology, China. Her research interests include Optimization Methods based on Swarm Intelligence, Deep Learning, Reinforcement Learning, and their Applications in Smart Cities and Intelligent Transportation. She has published over 100 papers, primarily in ACM\IEEE Transactions series journals and reputable conferences such as GECCO, NeurIPS, and ICLR. Dr. Gong was honored as the Pearl River Young Scholar by the Guangdong Education Department in 2017 and received the Guangdong Natural Science Funds for Distinguished Young Scholars in 2022.
Meta-Black-Box Optimization (MetaBBO) leverages a meta-level learner to automate the black-box optimization process, thereby enhancing efficiency and reducing human intervention. MetaBBO can be categorized into three distinct branches: the first focuses on automatic algorithm configuration or selection; the second employs neural networks to propose candidate solutions directly; and the third generates algorithms, as demonstrated in our study, by producing update rules expressed as closed-form equations. This talk provides a comprehensive overview of these three branches of MetaBBO, highlighting their methodologies, advantages, and current challenges.
Speaker: Setyo Tri Windras Mara, The University of New South Wales, Australia
Date: 2 September 2024
Time: 6:00 - 7:00pm (AET, UTC+10) [Convert to your local time]
Zoom link: https://vuw.zoom.us/j/96161300622
Setyo Tri Windras Mara is a PhD student at the University of New South Wales, Australia. His broad research interests include operational research (OR), transportation science, evolutionary optimizations, and healthcare management. To this day, his works have been published in various widely respected publishing avenues at the intersection between OR and computational intelligence. He also serves as a reviewer in several top journals, such as Computers & Operations Research, Computers & Industrial Engineering, and Expert Systems with Applications. In addition, Setyo is the co-initiator and a board member of OPSID, a community of young Indonesian researchers in OR and operations management.
Driven by the pursuit of sustainability, the integration of modern transportation modes such as electric vehicles (EVs) and drones has been a rising trend in logistics. Assessing the real-life implementation of such innovative systems requires significant capital investments. Thus, modelling and optimization approaches are preferred to assist the decision-making in this area. Still, challenges occur as the arising optimization models from the coordination between EVs and drones are computationally expensive to solve in practical problem sizes. Therefore, in my PhD research, evolutionary approaches are consulted to handle these challenges. In this talk, the effective strategies and lessons learned from the implementation of evolutionary approaches for solving EV-drone logistics models will be discussed.
If you have any questions or queries, please email Yi Mei or Fangfang Zhang.
Go to TOP
Go to TOP