JanMStraub / ML-bomberman-project Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Setup for a project/competition amongst students to train a winning Reinforcement Learning agent for the classic game Bomberman.

0 stars 0 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 76 Commits
agent_code		agent_code
assets		assets
logs		logs
replays		replays
results		results
screenshots		screenshots
.gitignore		.gitignore
Dockerfile		Dockerfile
README.md		README.md
agents.py		agents.py
environment.py		environment.py
events.py		events.py
fallbacks.py		fallbacks.py
items.py		items.py
main.py		main.py
replay.py		replay.py
settings.py		settings.py
test.py		test.py

Repository files navigation

Project "Bomberman Agent"

Project by Christian Teutsch and Jan Straub

Status

[] Work submitted (25.09.2023)
[] Report submitted (02.10.2023)

Project Rules

Develop at least two different models
Submit best-performing model to tournament
No restrictions on th techniques used
One model should use technique learned during semester
Written report at the end (10 pages p.p.)
Both models need to be described in the report

Game Rules

Four agents play the game
Each agent can move, drop bombs or stand still
Crates are destructable by bombs -> might reveal coins for additional points
Main objective is to eliminate other agents
No power-ups or special items

About

Setup for a project/competition amongst students to train a winning Reinforcement Learning agent for the classic game Bomberman.

Report repository

Releases

No releases published

Packages

No packages published

Contributors 2

Languages