Bio: Prof. C. L. Philip Chen is Dean and Professor at the School of Computer Science and Engineering, South China University of Technology. He is a Foreign Member of the European Academy of Sciences, a Member of the European Academy of Sciences and Arts, an IEEE Life Fellow, an IAPR Fellow, and an AAAS Fellow. His research interests include cybernetics, intelligent systems and control, computational intelligence, and data science. He has published more than 1,200 papers in leading international journals, including over 800 papers in IEEE Transactions journals, and has received numerous best paper awards. In terms of academic service, Prof. Chen served as President of the IEEE Systems, Man, and Cybernetics Society from 2012 to 2013, Editor-in-Chief of IEEE Transactions on Systems, Man, and Cybernetics: Systems from 2014 to 2019, and Editor-in-Chief of IEEE Transactions on Cybernetics from 2020 to 2022. His academic honors include the IEEE Norbert Wiener Award (2018), the IEEE Joseph G. Wohl Outstanding Career Award (2021), and the Wu Wenjun Outstanding Contribution Award in Artificial Intelligence. Through his research and academic leadership, Prof. Chen has made significant contributions to intelligent systems and control, computational intelligence, and artificial intelligence, while promoting advances in intelligent systems and related interdisciplinary fields.

Abstract: With the rapid development of large models, numerous research institutions worldwide have begun competing on parameter scale and computational power, prompting the academic community to critically reflect on the current trajectory of artificial intelligence. On the one hand, research and application of large models remain heavily dependent on a foreign-dominated high-performance computing ecosystem, making it difficult to ensure the autonomy and security of large-model technologies. On the other hand, most current industrial intelligence scenarios do not leverage large-model performance proportionally and incur higher computational costs; instead, more efficient and lightweight models are needed, particularly for edge deployment and open environments. More importantly, offline learning is currently lacking and urgently needed. To address this developmental challenge, this talk will integrate recent advances in both large and small models to explore breakthroughs in broad learning neural network computing, offline intelligence theory, and edge algorithm models.

Bio: Gary G. Yen received his Ph.D. degree in electrical and computer engineering from the University of Notre Dame in 1992. He was a Regents Professor in the School of Electrical and Computer Engineering, Oklahoma State University. He recently joined Sichuan University, College of Computer Science as a Chair Professor. His research interest includes intelligent control, computational intelligence, evolutionary multiobjective optimization, conditional health monitoring, signal processing and their industrial/defense applications. 

Gary was an associate editor of the IEEE Transactions on Neural Networks, IEEE Transactions on Evolutionary Computation, IEEE Transactions on Emerging Topics on Computational Intelligence, and IEEE Control Systems Magazine during 1994-1999, and of the IEEE Transactions on Control Systems Technology, IEEE Transactions on Systems, Man and Cybernetics (Parts A and B) and IFAC Journal on Automatica and Mechatronics during 2000-2010. He is currently serving as an associate editor for the IEEE Transactions on Cybernetics and IEEE Transactions on Artificial Intelligence.  Gary served as Vice President for the Technical Activities, IEEE Computational Intelligence Society in 2004-2005 and was the founding editor-in-chief of the IEEE Computational Intelligence Magazine, 2006-2009. He was elected to serve as the President of the IEEE Computational Intelligence Society in 2010-2011 and is elected as a Distinguished Lecturer for the term 2012-2014, 2016-2018, 2021-2023, and 2025-2027.  He received Regents Distinguished Research Award from OSU in 2009, 2011 Andrew P Sage Best Transactions Paper award from IEEE Systems, Man and Cybernetics Society, 2013 Meritorious Service award from IEEE Computational Intelligence Society and 2014 Lockheed Martin Aeronautics Excellence Teaching award. He is a Fellow of IEEE, IET and IAPR.

Abstract: Large-scale, dynamic multi-objective optimization problems (LSDMOPs) extend traditional DMOPs into high-dimensional decision spaces, reflecting the growing complexity of real-world dynamic systems under expected or unanticipated environmental changes. A typical example is the economic control of gas transportation in large-scale pipeline network, which involve hundreds or even thousands of decision variables, representing pipeline node pressures, compressor power settings, and valve operations. The system is simultaneously affected by various time-varying factors, such as consumer demand, market fluctuation, environmental temperature and humidity. In such contexts, algorithms are required to quickly find optimal solutions after each environment change, maintaining supply-demand balance while minimizing energy consumption. In practice, there are many more such examples, including dynamic resource allocation of 5G networks in dense urban environments, UAV swarm dispatch in a disaster relief scenario, and large-scale dynamic vehicle routing planning, just to name a few. However, the effectiveness of existing dynamic multi-objective evolutionary algorithms is severely limited for LSDMOPs, due to inadequate training data, predictions in unknown environments, and large-scale dynamic search spaces. To address these challenges, we propose a diffusion learning-based evolutionary framework, inspired by the intrinsic analogy between iterative evolution of optimization search and stepwise denoising in diffusion learning. Specifically, a new training paradigm is designed to learn the changing patterns of optimal regions in dynamic fitness landscapes. It achieves this by using populations’ evolutionary trajectories from initial solutions towards Pareto optimal solutions across historical environments as rich supervised training data. In addition, we introduce a trajectory alignment loss which encourages the stepwise denoising process to conform to the true population evolutionary behaviors in terms of spatial exploration, convergence trends, and boundary adaptation. The trained model can gradually control denoising direction and intensity using predefined conditions, allowing it to generate optimization paths from random noise toward Pareto-optimal solutions for a new environment. Along with an adversarial autoencoder-based large-scale dynamic multi-objective evolutionary framework and a large language model-based large-scale dynamic multi objective evolutionary framework, we will assess how deep generative modeling techniques and large-scale multi-objective evolutionary algorithms can be seamlessly integrated to solve large-scale DMOPs effectively and efficiently. Experimental results on a typical dynamic multi-objective test suite with problem settings from 10 to 1,000 dimensions demonstrate that the optimization performance of the proposed framework outperforms existing state-of-the-art designs. Especially in large-scale scenarios, the proposed framework is considered superior in terms of solution quality and computational efficiency.

Bio: Yang Tang (Fellow, IEEE) received the B.S. and Ph.D. degrees in electrical engineering from Donghua University, Shanghai, China, in 2006 and 2010, respectively. From 2008 to 2010, he was a Research Associate with The Hong Kong Polytechnic University, Hong Kong. From 2011 to 2015, he was a Post-Doctoral Researcher with the Humboldt University of Berlin, Berlin, Germany, and with the Potsdam Institute for Climate Impact Research, Potsdam, Germany. He is now a Professor with the East China University of Science and Technology, Shanghai. His current research interests include distributed estimation/control/optimization, agentic AI, computer vision, reinforcement learning, cyber-physical systems, hybrid dynamical systems, and their applications. Prof. Tang is an IEEE Fellow. He was a recipient of the Alexander von Humboldt Fellowship. He is a Co-Editor-in-Chief of IEEE Transactions on Industrial Informatics, an Senior Area Editor of IEEE Transactions on Circuits and Systems-I: Regular Papers, Associate Editor of IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on Cybernetics, IEEE Transactions on Industrial Informatics, IEEE/ASME Transactions on Mechatronics, IEEE Transactions on Circuits and Systems-I: Regular Papers, IEEE Transactions on Cognitive and Developmental Systems, IEEE Transactions on Emerging Topics in Computational Intelligence, IEEE Systems Journal, Engineering Applications of Artificial Intelligence (IFAC Journal), Science China Information Sciences, Science China Technological Sciences and Acta Automatica Sinica, etc. He is an IEEE Distinguished Lecturer from 2026-2027. He has published more than 200 papers in international journals and conferences, including more than 160 papers in IEEE Transactions and 20 papers in IFAC journals. He has been awarded as best/outstanding Associate Editor in IEEE journals for seven times. He is a (leading) guest editor for several special issues focusing on autonomous systems, robotics, and industrial intelligence in IEEE Transactions.

Abstract: The rapid development of embodied intelligence and large-scale foundation models is accelerating the transition of industrial systems toward Industry 5.0. This presentation first reviews the evolution of Industry 5.0 and embodied intelligence, examines the emerging industrial paradigm characterized by human–machine collaboration and autonomous intelligence, and systematically discusses the major forms and technical frameworks of embodied intelligence in the context of Industry 5.0 from three perspectives: individual intelligence, multi-agent intelligence, and collective intelligence. Considering the current stage of technological development, the presentation focuses on individual embodied intelligence, introducing its core architecture and fundamental operating principles. Building on this foundation, it highlights our team's recent research on embodied intelligence for industrial applications, with particular emphasis on embodied manipulation of objects made from challenging materials, as well as intelligent perception and manipulation in complex interactive environments. These studies explore key approaches to improving system robustness, environmental adaptability, and practical deployment capabilities in industrial settings. Finally, the presentation discusses the future trends and research directions of embodied intelligence from the perspective of Industry 5.0 across three levels: the model layer, the connectivity layer, and the application layer, providing insights into the large-scale adoption of embodied intelligence in next-generation industrial systems.

Bio: Liu Xinwang, Professor and Ph.D. Supervisor at the School of Computer Science, National University of Defense Technology (NUDT). His main research interests include machine learning and data mining. He was awarded the National Science Fund for Distinguished Young Scholars (2023) and the National Science Fund for Excellent Young Scholars (2019). He serves as the principal investigator for key projects of the National Natural Science Foundation of China (NSFC) and major projects under the "Science and Technology Innovation 2030" initiative, and is a core member of the NSFC Innovation Research Group (Category A). He has published over 150 papers in CCF Category A journals and conferences, including 20 ESI Highly Cited Papers. His total Google Scholar citations exceed 27,000, and he has been listed among the World's Top 2% Scientists for four consecutive years (2022–2024). He serves as an Associate Editor (AE) for top international journals such as IEEE T-KDE, IEEE T-NNLS, and IEEE T-CYB, as well as an Area Chair for premier international conferences including ICML and NeurIPS. His research achievements have been recognized with the First Prize of Natural Science Award from the China Computer Federation (2025), the First Prize of Wu Wenjun Natural Science Award from the Chinese Association for Artificial Intelligence (2024), the First Prize of Beijing Science and Technology Progress Award (2024), the First Prize of Hunan Provincial Natural Science Award (2014, 2021), the Second Prize of Hunan Provincial Natural Science Award (2025), and the Second Prize of Natural Science Award from the China Society of Image and Graphics (2024, 2025), among others.

Abstract: As a core technology for processing complex relational data, graph machine learning has been widely applied in social network analysis, recommendation systems, biomedicine, and other fields, demonstrating significant practical value. However, with the advancement of artificial intelligence research, performance is no longer the sole focus; whether a model is secure and reliable has become a crucial research direction for contemporary investigators. Through investigation, it is evident that existing graph learning methods are vulnerable to backdoor attacks and adversarial sample attacks targeting graph structures and attributes, leading to performance fluctuations and diminished reliability. To address these issues, our research group proposes the following innovative solutions: (1) a defense method based on self-supervised learning to counter backdoor attacks; (2) an importance-driven random sampling defense method to counter adversarial sample attacks. Furthermore, the group has conducted related research on knowledge graphs—a more specific and complex graph data type—providing guidance for achieving more secure graph learning. Finally, this report also reviews and outlines potential future research directions in graph learning.

Bio: Professor Zhixing Huang is now with the School of Computer Science and Engineering, South China University of Technology. He received his Bachelor's degree (2018) and Master’s degree (2020) from the same school. He was awarded a PhD in 2024 by the School of Engineering and Computer Science, Victoria University of Wellington, New Zealand. Prof. Zhixing Huang has nearly 30 published/accepted academic papers. He has won multiple top conference awards in the field of evolutionary computation, including Best Paper Awards and nominations at EuroGP 2022, GECCO 2023 and WCCI 2026. His main research interests lie in abstraction and reasoning capabilities of artificial intelligence, automatic program synthesis, knowledge discovery, and large-scale dynamic combinatorial optimization problems.

Abstract: Fitness landscape is an important perspective for analyzing and understanding search in genetic programming. However, due to the rugged and discrete nature of the program search space, existing research on the fitness landscape of genetic programming largely remains limited to difficulty analysis. In this talk, we go one step further in the study of genetic programming fitness landscapes. Specifically, we introduce approaches to optimize the fitness landscape of genetic programming to facilitate easier search, such as reducing local optima. Our experiments on symbolic regression and dynamic combinatorial optimization validate the potential of this new research direction.

Bio: Yi Jiang is currently an Assistant Professor with the School of Computer Science and Engineering, South China University of Technology, 510006 Guangzhou, China. He received the B.S. and Ph.D. degrees in computer science and technology from the South China University of Technology, Guangzhou, China, in 2020 and 2024, respectively. His research interests mainly include computational intelligence, evolutionary computation, machine learning, and their applications in real-world problems.

Abstract: Evolutionary multitask optimization seeks to improve search efficiency through knowledge transfer among optimization tasks. Identifying beneficial task relationships and preserving useful knowledge during transfer remain key challenges. This talk presents our research on relatedness identification and structured knowledge in evolutionary multitask optimization. We introduce a bi-objective framework that characterizes different aspects of task relatedness and selects appropriate transfer strategies, alongside a structure-preserving approach that acquires transferable knowledge and enables its direct and indirect propagation across tasks. The talk discusses how task relatedness and knowledge structure inform the design of effective transfer mechanisms for collaborative optimization.