Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
# Final Assignment

In this assignment, you will implement various model-based algorithms. You can choose to either continue working with the Catch environment or swtich to a new one.
In this assignment, you will either implement various policy gradient algorithms or model-based algorithms. You can choose to either continue working with the Catch environment or switch to a new one.

This template repository is almost the same as the previous one, with slight changes made to the Agent class.

If you decide to not use the Catch environment, make sure to adapt (or remove) the `environment.py` file.
If you decide not to use the Catch environment, make sure to adapt (or remove) the `environment.py` file.

## Installation

Expand All @@ -28,7 +28,7 @@ source test.sh

## Tips and Resources

Here are a couple of hints and resources that might help you with in this assignment:
Here are a couple of hints and resources that might help you with this assignment:

1. To help you out with technical writing, check out these papers for inspiration. Reading real scientific papers can help you out with using correct nomenclature and ensuring a clear structure. In particular, you can draw inspiration as to how complex concepts and formulas are introduced
and explained.
Expand All @@ -37,7 +37,7 @@ Here are a couple of hints and resources that might help you with in this assign

b. Paper summarising usage of RL in Chess - https://page.mi.fu-berlin.de/block/concibe2008.pdf

2. If you have duplicate code in multiple places, it’s probably a bad sign. Maybe you should try it to group that functionality in a seperate function?
2. If you have duplicate code in multiple places, it’s probably a bad sign. Maybe you should try to group that functionality in a separate function?
3. The agent should be able to learn using different types of algorithms. Maybe there is a way to make these algorithms easily swappable?
4. Type hinting is not required, but it can help your partner understand your code - https://mypy.readthedocs.io/en/stable/cheat_sheet_py3.html
5. Git workshop by Cover - https://studysupport.svcover.nl/?a=1
Expand Down