This project, developed for the Graphs and Complex Networks course, focuses on generating random graphs and applying advanced algorithms to determine the optimal path between a designated initial and final node. The project integrates well-established pathfinding techniques, such as A and Dijkstra, alongside reinforcement learning, a cutting-edge approach to solving complex decision-making problems.
The random graph generation ensures variability in the network structures, providing a diverse set of challenges for the algorithms to solve. A* and Dijkstra algorithms are used to compute the shortest path based on heuristic and cost functions, offering reliable and efficient solutions in deterministic scenarios. Reinforcement learning introduces an adaptive methodology, where an agent learns to navigate the graph by interacting with its environment, improving its performance over time.
This project demonstrates the practical application of graph theory and machine learning in network optimization and decision-making tasks. It provides insights into the strengths and limitations of different algorithms, showcasing their potential in solving real-world problems involving complex networks.
