Artificial intelligence (AI) makes it possible for machines to learn from experience, adjust to new inputs and perform human-like tasks. Most AI examples that you hear about today – from chess-playing computers to self-driving cars – rely heavily on deep learning and natural language processing.

A multi-agent system (MAS) is a computerized system composed of multiple interacting intelligent agents. Multi-agent systems can solve problems that are difficult or impossible for an individual agent or a monolithic system to solve.

Agent-based models are computer simulations used to study the interactions between people, things, places, and time. They are stochastic models built from the bottom up meaning individual agents (often people in epidemiology) are assigned certain attributes.

Computational Science is a rapidly growing multi- and interdisciplinary field. It develops mathematical and computational models and uses advanced computing techniques to simulate these models, driven by data.

Complex Network Analysis studies how to recognise, describe, visualise and analyse complex networks.

Network science is the study of connectivity and networks in all forms. From a single pair of nodes to complex networks with millions of members, network scientists map and analyze, identifying patterns between a network's attributes and their outcomes

Evolutionary computation is a sub-field of artificial intelligence (AI) and is used extensively in complex optimization problems and for continuous optimization. Evolutionary computation is used to solve problems that have too many variables for traditional algorithms.

Collective and swarm intelligence is the branch of artificial intelligence research that deals with systems with intelligent behavior consisting of a large number of actors. As such, it is a special part of multi-agent systems, where the role of individual agents is negligible and intelligent behavior appears at the community level. Individual agents can be cooperative (altruistic) or selfish.

Mechanism design is a special branch of game theory that deals with the creation of games (interaction protocols) that guide the collective behavior of several self-interested actors towards achieving a predetermined goal -- assuming rational agents.

Graph Neural Networks (GNNs) have emerged as a powerful technology with diverse applications. This research proposal goes beyond specific applications to explore the broader potential of GNNs. It focuses on three key aspects: robustness, explainability, and message passing. GNNs, while advanced, face challenges such as vulnerability to attacks and complex information handling. We aim to enhance their robustness against adversarial challenges, making them more reliable. Additionally, we aim to make GNNs more transparent and interpretable, bridging the gap between technology and human understanding. The research also delves into message passing techniques, crucial for efficient network analysis. By optimizing these techniques, we seek to improve network analysis methods. This proposal contributes to the broader field of GNN technology, unlocking new possibilities and empowering GNNs to excel in various applications, thereby enhancing knowledge discovery and network analysis.