GoBigger: A Scalable Platform for Cooperative-Competitive Multi-Agent Reinforcement Learning

GoBigger is a scalable platform aiming to delve into cooperative-competitive multi-agent reinforcement learning for swarm intelligence between multiple teams. GoBigger enables the simulation of various teams and agents in each team. This new game mode dramatically expands the way of agent cooperation and competition and can more effectively simulate the swarm intelligent agent behavior. GoBigger is also a complex game mainly focus on cooperation and competition between agents.
overview

Quick Start

Highlights

mid_eject

The clone balls eject towards the center and gather the size together.

teamwork

The clone balls split to eat the food balls.

merge

The clone balls split and merge.

eat_others

The clone balls split and eat other clone balls.