Facilitators Team

Dr. Dharshana Kasthurirathne Sri Lanka Institute of Information Technology (SLIIT) dharshana.k@sliit.lk	Dr. Mahima Weerasinghe Sri Lanka Institute of Information Technology (SLIIT) mahima.w@sliit.lk	Dr. Dinuka Sahabandu University of Washington (UoW) sdinuka@uw.edu
Mr. Jeewaka Perera Sri Lanka Institute of Information Technology (SLIIT) jeewaka.p@sliit.lk	Mr. Asiri Gawesha Sri Lanka Institute of Information Technology (SLIIT) asiri.l@sliit.lk	Mr. Sanka Mohottala Sri Lanka Institute of Information Technology (SLIIT) sanka.m@sliit.lk

[UPDATE - 14/02/25] We kindly ask you to fill in this anonymous form regarding the overall quality of the tutorial. As we are planning to present it to other venues, your feedback might be extremely important for us 🙏🙂

Abstract

Graph Neural Networks (GNNs) are revolutionizing the field of machine learning by enabling the analysis of graph-structured data, which is pervasive in domains such as social networks, molecular biology, and recommendation systems. Unlike traditional neural networks designed for grid-like data structures, GNNs utilize a message-passing mechanism, enabling nodes to iteratively aggregate information from their neighbors. This tutorial, titled ”Graph Neural Networks: From Zero to Hero,” introduces Sri Lankan researchers to the foundational concepts, advanced methodologies, and practical applications of GNNs.The tutorial spans eight modules, covering basics like graph theory and linear algebra, advanced GNN architectures such as Graph Convolutional Networks (GCNs) and Graph Attention Networks (GATs), and hands-on sessions using PyTorch Geometric (PyG). Applications explored include molecular property prediction, traffic forecasting, and computer vision tasks such as human action recognition. Special topics like graph expressivity, structural encoding, and challenges like oversmoothing are also ddressed. Participants will gain insights into cutting-edge developments like Geometric Deep Learning and Topological Neural Networks, paving the way for further research. This initiative is crucial, given the under-representation of GNN research in local conferences despite its global prominence. Through engaging, hands-on sessions leveraging Python and Google Colab, the tutorial caters to a diverse audience of students, academics, and industry professionals, requiring only foundational knowledge in deep learning and graph theory. This tutorial is organized by researchers from the BrAIN Labs Research Group at the Sri Lanka Institute of Information Technology (SLIIT).

GNN Tutorial Schedule

Time Duration	Description	Resource Person
08:30 AM - 08:35 AM	Introduction to tutorial	Dr. Dharshana Kasthurirathna
08:35 AM – 09:25 AM	Introduction to GNN and its applications Basic graph theory Graph neural networks Message passing GNN (MPGNN)	Dr. Mahima Weerasinghe
09:25 AM - 09:30 AM	Break
09:30 AM – 10:20 AM	GCN architecture GAT architecture	Mr. Sanka Mohottala
10:20 AM – 11:00 AM	GNN application (Coding session)	Mr. Asiri Gawesha
11:10 AM – 11:30 AM	New frontiers of GNN	Mr. Jeewaka Perera
11:30 AM – 11:40 AM	Q and A session	All resource persons will be present

Additional useful material

Facilitator Profiles

Dr. Dharshana Kasthurirathne

Dharshana Kasthurirathna is an academic and researcher with a Doctor of Philosophy from the Complex Systems Research Group at the University of Sydney, Australia, specializing in complex networks and evolutionary game theory. He also holds a master’s in computer science from the University of Colombo, and a BSc (Hons) in Computer Science and Engineering from the University of Moratuwa. Currently serving as an Assistant Professor at the Department of Computer Science, Faculty of Computing, Sri Lanka Institute of Information Technology (SLIIT), Dharshana’s research interests encompass evolutionary game theory, network science, machine learning, distributed systems, and optimization. In 2022, Dr.Kasthurirathna was the main presenter at the ”Graph Neural Networks: Theory and Applications” workshop, held as part of the 4th Inter- national Conference on Advancements in Computing (ICAC). He also obtained a SLIIT research grant in 2023 as the principle investigator for developing a lightweight graph neural network based human action recognition framework. He is also a member of BrAIN Labs research group as well as The Center of Excellence for AI (COE-AI) research group at SLIIT.

Dr. Mahima Weerasighe

Mahima Weerasighe obtained a first class honours in B.Eng. in Electronic Engineering from Sheffield Hallam University, U.K., in 2011, and the M.Sc. degree in applied electronics from the University of Colombo, Sri Lanka, in 2016. He completed his Ph.D. from Auckland University of Technology, New Zealand in 2023 on Neuromorphic Computing. He has also completed Research Assistantships in knowledge discovery research in predicting surgical outcomes and youth wellbe- ing, in New Zealand. He has published his research in multiple journals including Nature and IEEE. His research interests include using artificial intelligence techniques for human health and wellbeing. His current research interests include spiking neural networks, brain data processing, and computational neuroscience. He is a researcher of BrAIN Labs, a research and innovation consortium established to develop AI solutions for real-world challenges. He is also a member of the Center of Excellence for AI (COE-AI) research group at SLIIT

Dr. Dinuka Sahabandu

Dinuka Sahabandu is a Senior Lecturer (HG) at the Department of Computer Science, Faculty of Computing, Sri Lanka Institute of Information Technology (SLIIT). He received his Ph.D. from the Department of Electrical and Computer Engineering at the University of Washington, Seattle, WA, USA. He earned his B.S. and M.S. degrees in Electrical Engineering from Washington State University, Pullman, WA, USA, in 2013 and 2016, respectively. His research interests include the security and privacy of machine learning algorithms, applications of game theory and learning for cybersecurity, and the control of multi-agent networks. He also share an interest in the theoretical aspects of graph neural networks and the application of graph neural networks in his research domains. He is also a member of BrAIN Labs research group as well as the Center of Excellence for AI (COE-AI) research group at SLIIT.

Mr.Jeewaka Perera

Jeewaka Perera is a Senior Lecturer in the Department of Computer Science, Faculty of Computing, Sri Lanka Institute of Information Technology, Malabe, Sri Lanka. His research interests include multi-objective combinatorial optimization, graph neural networks, and reinforcement learning. His ongoing research includes Graph Neurual Networks, Graph Explainability and Biological AI. Mr. Perera received both his BSc. and MSc. in Computer Science from California State University, Fresno. He is also a member of BrAIN Labs research group as well as the Center of Excellence for AI (COE-AI) research group at SLIIT.

Mr. Sanka Mohottala

Sanka Mohottala is a Research Assistant and an MPhil candidate at Department of Computer Science, Faculty of Computing ,Sri Lanka Institute of Information Technology (SLIIT). He obtained his Bachelors in Electrical and Electronic Engineering in 2021 from SLIIT. His research interest are in Graph Neural Networks (GNNs), data-efficient deep learning methods and computer Vision. He has published in several conferences, including IEEE ICIT, TENCON and ICAC. He also served as a reviewer for multiple conferences and a IEEE Transaction Journal. He also conducted the graph neural networks related lectures in the Deep Learning module at SLIIT as a guest lecturer. He is experienced in multiple graph neural network related frameworks like PyTorch Geometric, DGL with hands-on experience in developing GNN-based models in multiple application domains. He co-conducted a workshop named ”Graph Neural Networks: Theory and Applications” at the 4th IEEE ICAC Conference. He also obtained a SLIIT research grant in 2023 as the a co-investigator for developing a lightweight graph neural network based human action recognition framework. He is also a member of BrAIN Labs research group as well as the Center of Excellence for AI (COE-AI) research group at SLIIT.

Mr. Asiri Gawesha

Asiri Gawesha is a Research Assistant and a MPhil candidate at the Department of Software Engineering, Faculty of Computing, Sri Lanka Institute of Information Technology (SLIIT). He previously contributing to a World Bank-funded project on the Culturally Sensitive Autism Assessment Tool (CSAAT). He holds a BSc in Electrical and Electronic Engineering from SLIIT, obtained in 2021. His research focuses on cutting-edge topics such as image processing, computer vision, edge computing, distributed computing, and cloud computing. His professional achievements include developing AI-powered autism screening tools, optimizing deep learning models for edge devices, and leading mobile app integrations for machine learning inference. He has published as an author at conferences like TENCON, CCWC and ICAC and furthermore, he has been involved in peer-review activities for IEEE conferences. He share an interest in graph neural network research areas like DeepSets and Transformers and have published work in those domains as well. He is also a member of BrAIN Labs research group as well as the Center of Excellence for AI (COE-AI) research group at SLIIT.

Details of the presenters of this tutorial

This tutorial is organized by the BrAIN Labs research group at SLIIT. All presenters are members of this research group

Name Mobile	Email Address
Dr. Mahima Weerasinghe	070462552	mahima.w@sliit.lk
Dr. Dharshana Kasthurirathna	0777671672	dharshana.k@sliit.lk
Dr. Dinuka Sahabandu	0777391621	dinuka.s@sliit.lk
Mr. Jeewaka Perera	0713771681	jeewaka.p@sliit.lk
Mr. Sanka Mohottala	0703975286	sanka.m@sliit.lk
Mr. Asiri Gawesha	0767616895	asiri.l@sliit.lk

Graph Neural Networks - From Zero to Hero

5th International Conference on Advanced Research in Computing (ICARC-2025)- Tutorials