New Regression Games Tools, Minetester, and HuggingFace Game Jam

Sharing updates from the EleutherAI team, Regression Games, and HuggingFace in the space of AI + games.

Welcome to our weekly newsletter and platform update! Every week we’ll bring you the latest AI + gaming news, resources, and updates from the Regression Games team.

Join our community on Discord and start playing!

Weekly Spotlight - Minetester from EleutherAI

The team over at EleutherAI published a blog post on their Minetester environment, an open-source and moddable voxel platform and RL tools for training agents. They talk about how research on top of projects such as MineDojo has been tricky due to the closed-source nature of Minecraft.

“… the entire stack is fully transparent, the codebase is smaller, and every dependency is open-source. This makes it easy for devs to dig into any component that could be causing an issue and lets us put our hooks directly into the Minetest game itself without unnecessary layers of indirection.”

This platform will likely make it easier to jump in and start training your own voxel-based RL agents, and we are excited to try it out! You can learn more about the project on their homepage, https://www.minetest.net/.

Regression Games Platform Update

As mentioned in our last update, the team has been working hard to get our initial bot platform ready for play testing. We have a ton of awesome features that we are excited to show off! If you have been looking for new AI-powered tools to decrease costs and time spent on testing games or building bots, we’d love to hear from you - please email [email protected] to get in touch.

Event Summary - HuggingFace AI Game Jam

Over the weekend, HuggingFace hosted a game jam where developers used open-source AI tools to build their games. Players used Stable Diffusion, AI NPCs, and other tools to make some pretty awesome games. With 80+ submissions, I think this was a huge success. The idea of a hackathon-style time to play with these tools is fun and a great learning experience - try playing some of the submitted games!

Technical Paper of the Week

This week, we take a look at a paper titled “RaidEnv: Exploring New Challenges in Automated Content Balancing for Boss Raid Games”. The authors explore the creation of a simulator for MMORPG raids, which can then be used to train agents and determine game balancing for different raid types. Automatic game balancing is an exciting application of AI game agents, and I’m excited to see where this research goes!

Members of our community discuss these papers on our Discord #reading-club channel. You can also find all past resources on our GitHub.

Looking for something new? Join the Regression Games team!

Have feedback about our newsletter? Let us know at [email protected]

You are receiving this email because you registered to our mailing list or are a player on our platform.