COMP90054 Assignment 3 Solution

You must read fully and carefully the assignment specification and instructions detailed in this file. You are NOT to modify this file in any way.
The aim of this project is to get you acquainted with value iteration and reinforcement learning, which is a good practise to have if you intended to do this for the pacman-contest.

Your task
You must build and submit your solution using the sample code we provide you in this repository, which is the same from the original UC
Berlkley code base. If you want to provide a report with your submission (e.g., reflections, acknowledgments, etc.), please do so in file
REPORT.md.
Please remember to complete the STUDENT.md file with your individual submission details (so we can identify you when it comes time to submit).
You should only work and modify files valueIterationAgents.py, qlearningAgents.py and analysis.py in doing your solution. Do not change the other Python files in this distribution.
Your code must not have carry any personal information, like your student number or your name. That info should go in the
STUDENT.md file, as per instructions above. If you use an IDE that inserts your name, student number, or username, you should disable that.
PacMan FAQ is available to answer common questions you might about the project on Piazza at https://piazza.com/class/kd49qfdh119p6?
cid=6
Getting started on GitHub - this video explains how to clone, git add, commit and push while developing your solution for the project: https://www.loom.com/share/ae7e93ab8bec40be96b638c49081e3d9
This assignment is worth 5% of your final mark (aditionaly).
Tasks
The details of each tasks are givin in the original UC Berlkley code base. Each task will help your understand more about the related subject content.
Marking criteria
Commit early, commit often: single or few commits with all the solution or big chucks of it, is not good practice.
Use meaningful commit messages: as a comment in your code, the message should clearly summarize what the commit is about.
Messages like "fix", "work", "commit", "changes" are poor and do not help us understand what was done.
Use atomic commits: avoid commits doing many things; let alone one commit solving many questions of the project. Each commit should be about one (little but interesting) thing.
Checking your submission
Run the following command to run sanity checks using our test files:
python ./autograder.py
NOTE: You should not change any files other than valueIterationAgents.py, qlearningAgents.py and analysis.py . You should not import any additional libraries into your code. This risks being incompatible with our marking scripts.
Submission Instructions
To submit your assignment you must complete the following four steps:
1. Complete the STUDENT.md file with your details of the submission.
2. Check that your solution runs on Python 3.6 and that your source code does not include personal information, like your student number or name.
3. Tag the commit version you want to be graded with tag submission.
Note that a tag is NOT a branch, so do not just create a branch called "submission" as that will not amount to tagging.
4. Fill the Project RL (Additional) Certification Form.
Please view the following to learn how to Tag your commit version you want to be graded:
How to create a Tag using the Command Line: https://www.loom.com/share/17ec72b954454bc89bbe1dbb0bd2874f
Another way to create a Tag using the User Interface: https://www.loom.com/share/3cd39e97919e4b688d9841613aba6973
Important information
Corrections: From time to time, students or staff find errors (e.g., typos, unclear instructions, etc.) in the assignment specification. In that case, corrected version of this file will be produced, announced, and distributed for you to commit and push into your repository. Because of that, you are NOT to modify this file in any way to avoid conflicts.
About this repo: You must ALWAYS keep your fork private and never share it with anybody in or outside the course, except your teammates, even after the course is completed. You are not allowed to make another repository copy outside the provided GitHub Classroom without the written permission of the teaching staff. Please respect the authors request:
Please do not distribute or post solutions to any of the projects.
Please remember to follow all the submission steps as per project specification.
GOOD LUCK and HAPPY PACMAN!
Acknowledgements