CZ3005 2020 Fall Assignment 2

Project Description

In this project, you need to implement one reinforcement learning algorithm (e.g., value iteration, policy iteration, Q-learning) for one grid-world-based environment: Treasure Hunting.

For more information, refer to the assignment document in this repository

Project setup

This project contains 2 files given to us:

1. environment.py - The environment used for this assignment
2. test.py - The file used to run the Agent in the environment

In order to run the Agent in the environment in test.py, the project requires these 2 files:

1. value_iteration_agent.py - The Agent to be let loose in the environment
2. utils.py - Utility functions that the Agent depends to perform additional computations in the environment.

If the project is not able to run due to missing dependencies, install the required packages manually or from requirements.txt by runing this command:

pip install -r requirements.txt