Bomberland

Overview

Bomberland is a multi-agent AI competition inspired by the classic console game Bomberman. Participants can build intelligent agents using strategies from tree search to deep reinforcement learning. The goal is to compete in a 2D grid world where players collect power-ups and place explosives to take down opponents. This repository contains the Bomberland engine and starter kits for working with the game engine.

Features

• Multi-agent AI competition inspired by Bomberman
• Intelligent agents can be built using strategies from tree search to deep reinforcement learning
• 2D grid world where players collect power-ups and place explosives
• Repo contains starter kits for working with the game engine + engine source
• Open AI gym wrapper for using the game engine as an Open AI gym environment

Summary

Bomberland is a multi-agent AI competition that allows participants to build intelligent agents using strategies ranging from tree search to deep reinforcement learning. The game engine provides a 2D grid world where players can collect power-ups and place explosives to take down opponents. The repository includes starter kits for different programming languages, as well as an Open AI gym wrapper for using the game engine in AI environments. Overall, Bomberland provides an exciting platform for AI developers to showcase their skills in a competitive setting.