Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature]: Make a Bomberman AI agent #6

Open
3 tasks
guilhem-sante opened this issue May 10, 2022 · 0 comments
Open
3 tasks

[Feature]: Make a Bomberman AI agent #6

guilhem-sante opened this issue May 10, 2022 · 0 comments
Labels
enhancement New feature or request

Comments

@guilhem-sante
Copy link
Contributor

guilhem-sante commented May 10, 2022

Description

The goal of this issue is to release an agent that can interact and learn (and hopely win) from Joep's Bomberman.

Expected behaviour

Ideally develop an agent with an advanced reinforcement learning algorithm (like rainbow) that make the win really hard for a real player. But an agent that can finish the game without killing himself is already something I think.

Environment

  • OS and version

Proposed solution

  • Update the possible actions from the new env (it will change the output of the model)
  • Update the processing of the new env states (it will change the input of the model)
  • Make such modifications modular. Nothing hardcoded or not simple to adjust to any env.
@guilhem-sante guilhem-sante added enhancement New feature or request triage Need to be tagged labels May 10, 2022
@staeter staeter removed the triage Need to be tagged label Jun 15, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

When branches are created from issues, their pull requests are automatically linked.

8 participants