Do you remember playing the thrilling game of Feeding Frenzy? Get ready to experience a Decision Intelligence twist with the AI BOB: Go-Bigger Multi-Agent Decision Intelligence Challenge! This exciting competition invites technology developers and students worldwide to build intelligent agents that can outsmart enemies and gain maximum scores.
Challenge Introduction
The GoBigger Challenge focuses on multi-agent confrontation, an essential area in decision intelligence AI. Developed by OpenDILab, the GoBigger game aims to foster research and talent development in this field, creating a leading, original, and open-source ecosystem for decision intelligence AI. At the end of the competition, all agents will be rigorously evaluated, and rankings will be established according to performance.
Tasks in the Challenge
Your main task in this challenge is to design an intelligent agent to control a team in the Go-Bigger game environment. Each team consists of multiple players, and the goal is to utilize teamwork effectively to achieve the highest score possible.
How to Make Your Submission
To participate, you’ll need to implement your own submission based on provided examples from the submission repository. Below is a simple explanation of the submission process, with the help of an analogy:
- Team Name and Player Names: Imagine your team as a group of superheroes, each with unique capabilities. You need to introduce your superhero team by providing a team name and a list of heroes (player names).
- The Action Plan: Just like superheroes need to decide their moves, you must implement the get_actions function to define how your agents will react based on the current game situation.
- Observing the Playfield: Your agents will receive an ‘observation’ of the game, just like superheroes scouting the battlefield. This observation will help them make decisions.
class BaseSubmission:
def __init__(self, team_name, player_names):
self.team_name = team_name
self.player_names = player_names
def get_actions(self, obs):
raise NotImplementedError
Example Submissions
The challenge provides examples like RandomSubmission and BotSubmission, illustrating how different approaches can be employed. Utilize these examples to guide your implementation.
Try Your First Submission
If you’re feeling overwhelmed, don’t fret! You can quickly generate a basic submission using our simplest case submission. Here’s how:
$ cd submitsubmission_example
$ python -u test.py
This will check for correctness and notify you if your submission is successful. Once you’re done, just upload the generated my_submission.tar.gz.
Troubleshooting and Support
In case you run into issues, here are some helpful tips:
- Ensure that your project structure mirrors what is described in the submission section. Every file must be in the right directory.
- Double-check the implementation of the get_actions function; it must return the expected results based on the provided observations.
- If you encounter difficulties, visit the relevant links or consult the documentation for additional guidance.
- For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.
At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.
Resources
For additional guidance, you can explore the following resources:
We are looking forward to your brilliant performance in this challenge! Put on your superhero cape and get started!