Francesco Agnelli | Graph Neural Networks | Best Researcher Award

Posted on by sciencefather

Dr. Francesco Agnelli | Graph Neural Networks | Best Researcher Award

Dr. Francesco Agnelli, University of Milan, Italy

Francesco Agnelli is an Italian researcher and PhD candidate at the University of Milan, where he delves into the frontiers of deep learning and artificial intelligence. Born in 1998 in CantΓΉ, Italy, Francesco began his academic journey with a stellar Bachelor’s and Master’s degree in Mathematics, both earned cum laude at the University of Insubria. His academic focus evolved from Morse Theory to applied mathematics and computational intelligence, leading to his cutting-edge research on Graph Neural Networks (GNNs). Francesco's work bridges advanced machine learning methods with real-world problems like affective computing and graph isomorphism. With industry experience at Power Reply and teaching stints in local schools, he combines theory with practical impact. Francesco also shares his knowledge as a tutor and speaker in major academic and tech platforms, including an NVIDIA webinar and the 2024 ECCV Conference. He continues to contribute to the intersection of mathematics, AI, and neural computation.

Professional Profile 

Orcid

Summary of Suitability for the 'Research for Best Researcher Award'

Francesco Agnelli stands out as an exceptional candidate for the 'Research for Best Researcher Award' due to his impressive academic background, comprehensive research experience, and contributions to cutting-edge fields in mathematics and computer science, particularly through his work with Graph Neural Networks (GNNs).

Francesco holds a Master’s degree in Mathematics with honors (110/110 cum laude) from UniversitΓ  degli Studi dell'Insubria, where his academic pursuits focused on computational mathematics and machine learning. His work has specifically contributed to the application of GNNs in the domains of graph isomorphism and affective computing, demonstrating his ability to innovate within highly specialized research areas. His ongoing Ph.D. studies at UniversitΓ  degli studi di Milano further highlight his dedication to advancing human understanding through deep learning and multi-modal input integration.

Education 

Francesco Agnelli holds a Master’s degree in Mathematics with highest honors (110/110 cum laude) from the University of Insubria, where he focused on computational mathematics and machine learning. His thesis explored the application of Graph Neural Networks to the graph isomorphism problem. During his studies, he completed an Erasmus semester at KU Leuven, Belgium, earning top grades in advanced courses like Wavelets and Applications (29/30), Life Insurance (30/30), and Machine Learning (29/30). He also earned a Bachelor’s degree in Mathematics, again with 110/110 cum laude, from the same university. Francesco participated in a 24 CFU course for teacher training and holds a Scientific High School diploma from Liceo Scientifico Enrico Fermi with a score of 96/100. Now pursuing a PhD at the University of Milan, Francesco investigates the use of GNNs in affective computing, incorporating multimodal inputs and foundation models to advance human-centered AI.

Professional Experience

Francesco Agnelli is currently a PhD student at the University of Milan, conducting research at PhuseLab on deep learning and human understanding. His work integrates Graph Neural Networks and foundation models to process multimodal affective data. Previously, Francesco worked as an IT Consultant at Power Reply in Milan, where he supported the Eni Multicard CRM project. His tasks included transitioning the system from Siebel to Salesforce, performing data analysis, and implementing technical corrections using tools like SOQL and Excel. Earlier, he served as a Mathematics and Science teacher at Istituto Comprensivo Como Lora, where he engaged with younger students to foster scientific curiosity. In parallel, he has tutored university-level courses like Mathematical Analysis and Computational Mathematics. His technical fluency spans Python (especially PyTorch), Matlab, SOQL, and Java. Francesco’s diverse experiences reflect a strong ability to bridge academic rigor with real-world application in both corporate and educational environments.

Awards and Recognition

Francesco Agnelli has received consistent recognition throughout his academic and professional journey. He graduated cum laude in both his Bachelor’s and Master’s degrees in Mathematics, reflecting outstanding academic excellence. As a department representative and member of multiple university commissions at the University of Insubria, Francesco was honored for his leadership and advocacy in academic governance. He was also selected as a university tutor, mentoring students in Mathematical Analysis and Computational Mathematics. His expertise in artificial intelligence earned him an invitation to speak at the prestigious NVIDIA webinar on "Enhancing Visual Understanding With Generative AI". Furthermore, he was a volunteer at the renowned ECCV 2024 Conference, showcasing his commitment to engaging with the AI research community. These accolades affirm his growing impact in academia and the AI research landscape, positioning him as a promising thought leader in the field of Graph Neural Networks and beyond.

Research Skill On Graph Neural Networks

Francesco Agnelli’s research skills are deeply rooted in computational mathematics, with a specialized focus on Graph Neural Networks (GNNs). His academic evolutionβ€”from Morse Theory to graph isomorphism problemsβ€”demonstrates a solid foundation in abstract mathematics and its translation into real-world computing tasks. In his PhD at the University of Milan, he explores the fusion of GNNs with affective computing, particularly multimodal input processing and fine-tuning of foundation models. His technical toolkit includes Python (PyTorch), Matlab, and data query languages like SOQL. Francesco exhibits high proficiency in integrating theoretical algorithms with deep learning frameworks, often experimenting with cross-domain solutions. He has hands-on experience working with numerical analysis, approximation methods, and neural architectures, allowing him to simulate and interpret graph-structured data effectively. His research reflects a forward-thinking and collaborative approach to AI that bridges data, emotions, and decision-making through intelligent systems grounded in strong mathematical logic.

  Publication Top Notes

  • Title: KA-GCN: Kernel-Attentive Graph Convolutional Network for 3D face analysis

  • Authors: Francesco Agnelli, Giuseppe Facchi, Giuliano Grossi, Raffaella Lanzarotti

  • Journal: Array

  • DOI: 10.1016/j.array.2025.100392

  • Year: 2025

  • Citation: Agnelli, F., Facchi, G., Grossi, G., & Lanzarotti, R. (2025). KA-GCN: Kernel-Attentive Graph Convolutional Network for 3D face analysis. Array. https://doi.org/10.1016/j.array.2025.100392

Yukun Shi | Computer Science | Best Scholar Award

Posted on by sciencefather

Assoc. Prof. Dr. Yukun Shi | Computer Science | Best Scholar Award

Assoc. Prof. Dr. Yukun Shi, Beijing University of Chemical Technology, China

Dr. Yukun Shi is an accomplished researcher and Associate Professor at the Department of Information Science and Technology, Beijing University of Chemical Technology. He specializes in multi-agent systems, control system network attacks, and distributed estimation. Dr. Shi earned his Ph.D. in Control Science and Engineering from Beijing University of Chemical Technology in 2022. His academic journey includes a one-year research visit to the University of Victoria, Canada, in 2021. His contributions to the field are significant, particularly in advancing secure state estimation and consensus control. He has published extensively in top-tier journals, addressing challenges in network security and distributed control. With a strong background in system modeling and cybersecurity, Dr. Shi continues to drive innovations in multi-agent collaboration and resilience against malicious attacks. His research not only contributes to theoretical advancements but also has practical implications for industrial and technological applications worldwide.

Professional Profile

Orcid

Suitability for the Research for Best Scholar Award – Yukun Shi

Dr. Yukun Shi, an Associate Professor at the Beijing University of Chemical Technology, has demonstrated remarkable academic and research excellence in the field of control science and engineering. His expertise spans critical areas such as multi-agent systems, control system network attacks, distributed estimation, and consensus control, making his contributions highly relevant to modern automation and cybersecurity challenges. His work is particularly notable in the area of secure state estimation, where he has investigated the robustness of networked control systems against malicious sensor attacks, an emerging concern in industrial and cyber-physical systems.

Dr. Shi’s research output includes several publications in prestigious IEEE journals, such as IEEE Transactions on Automation Science and Engineering and IEEE Transactions on Control of Network Systems, highlighting his ability to contribute cutting-edge advancements in his field. His scholarly work is well-cited, reflecting both its impact and recognition within the scientific community. Additionally, his international exposure, including a research visit at the University of Victoria, Canada, underscores his global perspective and collaborative research approach.

πŸŽ“ Education 

Dr. Yukun Shi pursued his Ph.D. in Control Science and Engineering at Beijing University of Chemical Technology, graduating in 2022. His doctoral research focused on secure state estimation in multi-agent systems under adversarial conditions, bridging control theory with cybersecurity. As part of his academic development, he undertook a one-year research visit to the University of Victoria, Canada, in 2021, where he collaborated on cutting-edge projects related to network security and control systems. His education provided him with a strong foundation in distributed control, estimation algorithms, and robust filtering techniques. Throughout his studies, Dr. Shi honed his expertise in tackling cyber threats to industrial control systems, laying the groundwork for his future research in resilient multi-agent networks. His academic journey is marked by rigorous training, innovative problem-solving, and contributions to the field of control and automation engineering.

πŸ’Ό Professional Experience

Dr. Yukun Shi currently serves as an Associate Professor at the Department of Information Science and Technology, Beijing University of Chemical Technology. With a research focus on multi-agent systems, network security, and distributed estimation, he has made significant contributions to securing cyber-physical systems. His professional journey includes leading research projects on sensor attacks, consensus control, and fault-tolerant filtering in distributed networks. Dr. Shi actively collaborates with international institutions to develop advanced methodologies for improving the resilience of control systems against malicious threats. His role extends beyond research, encompassing mentorship, curriculum development, and industry partnerships. He is a sought-after speaker at academic conferences and has peer-reviewed numerous articles in high-impact journals. His dedication to cybersecurity and control engineering has positioned him as a thought leader in the field, driving innovation and practical solutions to safeguard modern industrial and technological infrastructures.

πŸ… Awards and Recognition 

Dr. Yukun Shi has received multiple accolades for his pioneering work in control systems and cybersecurity. He has been recognized for his contributions to secure multi-agent systems and networked control security. His research papers have been published in high-impact journals, earning him best paper awards at leading automation and control conferences. Dr. Shi has also been a recipient of prestigious research grants that support his work in developing robust estimation algorithms against cyber threats. His outstanding contributions have been acknowledged by industry associations, positioning him as a key figure in distributed system security. His work has not only influenced academia but also guided practical implementations in industrial automation and cybersecurity frameworks. Additionally, Dr. Shi has served as a reviewer for top-tier journals, further highlighting his expertise and influence in the scientific community. His relentless pursuit of excellence continues to shape the future of secure control systems.

🌍 Research Skills On Computer Science

Dr. Yukun Shi possesses a robust research skill set centered around multi-agent systems, control system security, and distributed estimation. His expertise includes developing secure state estimation techniques to mitigate network attacks in cyber-physical systems. He specializes in designing fault-tolerant control algorithms that enhance the resilience of distributed networks. His research also encompasses consensus control strategies to improve synchronization in multi-agent environments. Dr. Shi is proficient in advanced filtering techniques, such as Kalman filtering and observer design, to ensure accurate system monitoring despite adversarial interference. He actively applies mathematical modeling and optimization methods to enhance decision-making in complex systems. His work in secure control frameworks has broad applications in autonomous systems, industrial automation, and networked infrastructures. With a keen focus on practical implementation, Dr. Shi’s research continues to bridge theoretical advancements with real-world security challenges, contributing to the evolution of resilient cyber-physical networks.

πŸ“– Publication Top Notes

  • Title: Optimal Output-Feedback Controller Design Using Adaptive Dynamic Programming: A Permanent Magnet Synchronous Motor Application
    • Authors: Zhongyang Wang, Huiru Ye, Youqing Wang, Yukun Shi, Li Liang
    • Citation: IEEE Transactions on Circuits and Systems II: Express Briefs
    • Year: 2025
  • Title: Distributed Filter Under Homologous Sensor Attack and Its Application in GPS Meaconing Attack
    • Authors: Yukun Shi, Wenjing He, Li Liang, Youqing Wang
    • Citation: IEEE Transactions on Automation Science and Engineering
    • Year: 2024
  • Title: Event-triggered distributed secure state estimation for homologous sensor attacks
    • Authors: Yukun Shi, Haixin Ma, Jianyong Tuo, Youqing Wang
    • Citation: ISA Transactions
    • Year: 2023
  • Title: Distributed Secure State Estimation of Multi-Agent Systems Under Homologous Sensor Attacks
    • Authors: Yukun Shi, Youqing Wang, Jianyong Tuo
    • Citation: IEEE/CAA Journal of Automatica Sinica
    • Year: 2023
  • Title: Online Secure State Estimation of Multiagent Systems Using Average Consensus
    • Authors: Yukun Shi, Youqing Wang
    • Citation: IEEE Transactions on Systems, Man, and Cybernetics: Systems
    • Year: 2022
  • Title: Asymptotically Stable Filter for MVU Estimation of States and Homologous Unknown Inputs in Heterogeneous Multiagent Systems
    • Authors: Yukun Shi, Changqing Liu, Youqing Wang
    • Citation: IEEE Transactions on Automation Science and Engineering
    • Year: 2022
  • Title: Secure State Estimation of Multiagent Systems With Homologous Attacks Using Average Consensus
    • Authors: Yukun Shi, Changqing Liu, Youqing Wang
    • Citation: IEEE Transactions on Control of Network Systems
    • Year: 2021