This repository is created to implement the basics of reinforcement learning by determining a policy via value iteration. The assignment also uses a specialized form of artificial neural networks (called Self Organizing Maps) for clustering and visualization.
A value iteration agent that navigates in a n * n grid is implemented and visualized. The grid world offers reward (+100) at different points. The agent is moving around (via Up, Down, Left, Right actions) to maximize its reward. A policy is determined via value iteration that guides the agent while navigating in the grid. There are certain obstacles in the grid that restrict the movement of agent.
Self-organizing maps is applied on the timeseries dataset of covid cases and clustering is visualized