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

Sai Venkatesh Chilukoti | Computer Science | Best Researcher Award

Posted on by sciencefather

Mr. Sai Venkatesh Chilukoti | Computer Science | Best Researcher Award

Mr. Sai Venkatesh Chilukoti, University of Louisiana at Lafayette, United States

Sai Venkatesh Chilukoti is a dedicated researcher in Computer Engineering, currently pursuing a Ph.D. at the University of Louisiana at Lafayette under Dr. Xiali Hei. With a stellar academic record and a CGPA of 4.0, he specializes in Deep Learning, Network Security, and Cyber-Physical Systems. His research interests span Federated Learning, Differential Privacy, and Machine Learning applications in healthcare and security. Sai Venkatesh has contributed to multiple peer-reviewed journals and conferences, focusing on privacy-preserving AI and identity recognition. As a Research Assistant in the Wireless Embedded Device Security (WEDS) Lab, he has worked on cutting-edge projects integrating AI, privacy-enhancing techniques, and embedded security. With experience as a Teaching Assistant, he mentors students in Neural Networks, Python, and AI-related fields. His passion lies in translating research into real-world applications, particularly in medical imaging and cybersecurity. Sai is fluent in English, Hindi, and Telugu, and has strong technical skills in Python, PyTorch, and TensorFlow.

Professional Profile

Scopus

Orcid

Google Scholar

Suitability for the Research for Best Researcher Award – Sai Venkatesh Chilukoti

Sai Venkatesh Chilukoti demonstrates an impressive academic and research portfolio, making him a strong contender for the Research for Best Researcher Award. Currently pursuing a Ph.D. in Computer Engineering at the University of Louisiana at Lafayette with a perfect 4.0/4.0 CGPA, he has developed expertise in deep learning, distributed computing, network security, and cyber-physical systems. His academic credentials are further strengthened by a solid foundation in electronics and communication engineering at the undergraduate level.

His research contributions are notable, particularly in privacy-preserving deep learning, federated learning, and medical AI applications. His work on diabetic retinopathy classification, gastrointestinal cancer prediction, and differential privacy models for healthcare data showcases both technical depth and real-world impact. His involvement in cutting-edge machine learning techniques, including LSTMs, Transformers, and convolutional networks, highlights his ability to innovate within the field. Furthermore, his research assistantship in Wireless Embedded Device Security (WEDS) Lab and role as a teaching assistant demonstrate both research rigor and mentorship capabilities.

πŸŽ“ Education 

Sai Venkatesh Chilukoti is currently pursuing a Ph.D. in Computer Engineering at the University of Louisiana at Lafayette (2021-2025, expected), under the supervision of Dr. Xiali Hei, with a perfect CGPA of 4.0. His coursework includes Deep Learning, Network Security, Distributed Computing, and Cyber-Physical Systems. His research focuses on privacy-preserving AI, federated learning, and deep learning model optimization.

He completed his B.Tech. in Electronics and Communication Engineering at Velagapudi Ramakrishna Siddhartha Engineering College (2017-2021), earning a CGPA of 8.53/10. His undergraduate studies encompassed AI, Python, Artificial Neural Networks, and Digital Signal Processing.

Throughout his education, Sai has actively engaged in research projects, including identity recognition using mmWave radar sensors, privacy-aware medical imaging, and deep learning applications in cybersecurity. His academic journey reflects a strong foundation in computational intelligence and a commitment to solving real-world challenges through innovative AI techniques.

πŸ’Ό Professional Experience

Sai Venkatesh Chilukoti has extensive research and teaching experience, specializing in Deep Learning, Cybersecurity, and Federated Learning. As a Research Assistant at the Wireless Embedded Device Security (WEDS) Lab (2021-present), he has worked on privacy-preserving AI models, security solutions for embedded devices, and deep learning-based medical imaging applications. His work includes designing federated learning frameworks for decentralized AI and developing privacy-aware deep learning techniques.

As a Teaching Assistant for Neural Networks (2024-present), Sai mentors students in probability, calculus, and AI programming using PyTorch and Scikit-learn.

He has led numerous projects, such as statistical analysis of COVID-19 data, AI-driven financial forecasting, and deep learning applications in 3D printing quality control. His expertise extends to programming in Python, C, SQL, and MATLAB, along with experience in cloud computing and AI model deployment. He has also reviewed papers for IEEE Access and other reputed journals.

πŸ… Awards and Recognition 

Sai Venkatesh Chilukoti has been recognized for his outstanding contributions to AI research, deep learning, and cybersecurity. He has received multiple conference paper acceptances, including at the Hawai’i International Conference on System Sciences (HICSS-56) and CHSN2021. His work on privacy-preserving AI has been published in high-impact journals like BMC Medical Informatics and Decision Making and Electronic Commerce Research and Applications.

He has also earned certifications in Deep Learning Specialization (Coursera), AI for Everyone (deeplearning.ai), and Applied Machine Learning in Python (University of Michigan). His research in Federated Learning has gained attention for its innovative approach to privacy protection in healthcare AI models. Additionally, Sai has contributed as a reviewer for IEEE Access and Euro S&P, demonstrating his expertise in computer security and AI ethics. His contributions to machine learning, cybersecurity, and privacy-aware AI continue to impact both academic and industrial domains.

🌍 Research Skills On Computer Science

Sai Venkatesh Chilukoti specializes in Federated Learning, Differential Privacy, and Deep Learning model optimization. His expertise spans AI-driven cybersecurity, identity recognition using mmWave radar sensors, and privacy-preserving medical imaging. He has worked extensively with machine learning frameworks such as PyTorch, TensorFlow, and Scikit-learn.

Sai has developed AI models for secure collaborative learning, utilizing techniques like DP-SGD for privacy preservation. His research also explores transformer-based architectures, convolutional networks, and ensemble learning methods to enhance predictive performance. He has integrated advanced optimization techniques, including adaptive gradient clipping and label smoothing, into deep learning pipelines.

He has hands-on experience with federated learning platforms like Flower and privacy-preserving AI models in medical data analysis. His work in statistical modeling, computer vision, and neural networks has contributed to breakthroughs in security and healthcare AI. Sai’s research aims to advance AI applications while maintaining ethical and privacy standards.

πŸ“– Publication Top Notes

  • A reliable diabetic retinopathy grading via transfer learning and ensemble learning with quadratic weighted kappa metric
      • Authors: Sai Venkatesh Chilukoti, Liqun Shan, Vijay Srinivas Tida, Anthony S. Maida, Xiali Hei
      • Journal: BMC Medical Informatics and Decision Making
      • Volume: 24, Issue 1
      • Article Number: 37
      • Year: 2024
  • Privacy-Preserving Deep Learning Model for Covid-19 Disease Detection
      • Authors: Vijay Srinivas Tida, Sai Venkatesh Chilukoti, Sonya H. Y. Hsu, Xiali Hei
      • Conference: 56th Hawaii International Conference on System Sciences
      • Year: 2023
  • Single Image Multi-Scale Enhancement for Rock Micro-CT Super-Resolution Using Residual U-Net
    • Authors: Liqun Shan, Chengqian Liu, Yanchang Liu, Yazhou Tu, Sai Venkatesh Chilukoti
    • Journal: Applied Computing and Geosciences
    • Year: 2024
  • Kernel-Segregated Transpose Convolution Operation
    • Authors: Vijay Srinivas Tida, Sai Venkatesh Chilukoti, Sonya H. Y. Hsu
    • Conference: 56th Hawaii International Conference on System Sciences
    • Year: 2023
  • Modified ResNet Model for MSI and MSS Classification of Gastrointestinal Cancer
    • Authors: Sai Venkatesh Chilukoti, C. Meriga, M. Geethika, T. Lakshmi Gayatri, V. Aruna
    • Book Title: High Performance Computing and Networking: Select Proceedings of CHSN 2021
    • Year: 2022
  • Enhancing Unsupervised Rock CT Image Super-Resolution with Non-Local Attention
    • Authors: Chengqian Liu, Yanchang Liu, Liqun Shan, Sai Venkatesh Chilukoti, Xiali Hei
    • Journal: Geoenergy Science and Engineering
    • Volume: 238
    • Article Number: 212912
    • Year: 2024
  • Method for Performing Transpose Convolution Operations in a Neural Network
    • Inventors: Vijay Srinivas Tida, Sonya Hsu, Xiali Hei, Sai Venkatesh Chilukoti, Yazhou Tu
    • Patent Application: US Patent App. 18/744,260
    • Year: 2024
  • IdentityKD: Identity-wise Cross-modal Knowledge Distillation for Person Recognition via mmWave Radar Sensors
    • Authors: Liqun Shan, Rujun Zhang, Sai Venkatesh Chilukoti, Xingli Zhang, Insup Lee
    • Conference: ACM Multimedia Asia
    • Year: 2024
  • Facebook Report on Privacy of fNIRS Data
    • Authors: M. I. Hossen, Sai Venkatesh Chilukoti, Liqun Shan, Vijay Srinivas Tida, Xiali Hei
    • Preprint: arXiv preprint arXiv:2401.00973
    • Year: 2024