Spiderman Agent Learns to Swing and Intro to RLHF

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 - GitHub Repositories for Learning

Getting started with AI in games can be tricky, with so many topics and concepts to learn. This week, we wanted to highlight two repositories on GitHub with resources to learn and get started.

awesome-ai-tools-for-game-dev from Thomas Simonini

This repository from Thomas Simonini, a developer advocate from HuggingFace, has tons of great resources for using AI in games, whether it be asset generation, voice generation, AI agents, and more.

Gaming-AI-and-RL from Regression Games

This repository from Regression Games covers guides, papers, tools, and tutorials for AI agents in games. It spans topics in reinforcement learning, classical AI, and how to have fun with these tools.

Channel Highlight - b2studios

Whether it’s training agents to swing like Spiderman, or finding interesting strategies to get a strike in bowling, the b2studios channel is a great beginner resource to learn more about reinforcement learning and AI agents in games. As an added bonus, their videos are hilarious.

Blog Post - Illustrating Reinforcement Learning from Human Feedback (HuggingFace)

If you’ve been staying up to date in the world of reinforcement learning, you’ll know that RLHF is a hot topic. This blog post from HuggingFace gives a great explanation of what it is, how it works, and the tools that are available to try it out. We definitely recommend reading this to get up to speed!

Twitter / Thread Highlight

Mikayel Samvelyan is a PhD candidate at UCL and FAIR who tweets frequently about advancements and ideas in Multi-Agent RL. In addition to giving them a follow, this thread on MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning is quite interesting!

In the News and Upcoming Events

The 22nd International Conference on Autonomous Agents and Multiagent Systems is coming up at the end of the month - keep an eye on this!

Technical Paper of the Week

This week, we will be taking a look at a recently-published paper called RAMario: Experimental Approach to Reptile Algorithm Reinforcement Learning for Mario. This paper explores using the Reptile Algorithm against other approaches such as PPO and DQN in Mario gameplay tasks.

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.