New LoL Bots and Game AI Learning Prompts in ChatGPT

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 - Riot Games Blog Post on Bots

The developer team at Riot Games released a blog post last week detailing their efforts and endeavors around bots on their platform. In the past, League of Legends (LoL) bots have been quite limited, both in strategies and character selection. Their blog post details new plans for better tech (such as enhanced Behavior Trees and ML) that will be shared between their internal teams to make bots that can work together, cover more champion types, and provide a better experience for new players.

Regression Games Platform Updates

Last week, we published a blog post reflecting on our recent Alpha Cup tournament. We discuss some of the main learnings and what we plan to work on next.

The team has several platform updates coming in the near future:

• A new way to develop bots outside of Git (for players who don’t want to use Git or GitHub).

• New interfaces for programming bots, such as behavior trees or GOAP interfaces, with generative AI features.

• A Unity game engine prototype for new AI games and partnerships.

ChatGPT Learning Prompts

As you learn more about AI in games, you’ll find that ChatGPT serves as a great way to dive deeper into topics and discover details that articles or tutorials may not cover.

Each week, we will give a prompt to explore and learn more about AI in games. Let’s try this prompt:

Explain GOAP to me, assuming I am new to game AI programming. Provide a simple example.

Dive deeper by following up with prompts like this:

• What are the downsides to using GOAP?

• Can you provide a Python implementation of GOAP?

• What are the best frameworks or tutorials for learning GOAP?

If you come up with other prompts or results that you find interesting, tweet at us @RegressionGG!

Technical Paper of the Week

The International Conference on Learning Representations took place last week. Papers covered topics in LLMs, reinforcement learning, and games.

The paper Emergence of Maps in the Memories of Blind Navigation Agents tests whether blind AI agents can learn and memorize an environment for navigation. The paper shows that these agents learn representations of their environment, and they even perform well in new environments!

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.