- The Regression Games Weekly Observation
- Posts
- Unity + HuggingFace and Self Driving Cars
Unity + HuggingFace and Self Driving Cars
This week we take a look at a new partnership between HuggingFace and Unity.
Aaron Vontell
June 13, 2023
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.
Regression Games is hiring! Join our team to work on exciting projects at the intersection of AI and gaming. We have great benefits, competitive pay, and an amazing team. 👉 https://regression.gg/careers
Weekly Spotlight - HuggingFace + ML-Agents
Last week, HuggingFace announced that it has officially migrated its ML-Agent HuggingFace Integration (e.g. the HuggingFace Hub) into the official Unity repository, making it easier for users of these platforms to build RL-based agents for Unity games. We are really excited to check out the new course notes and try out these tutorials!
Deep Dive - Key RL Papers
More reinforcement learning papers! Sometimes it can be hard to determine where to start with a complex topic like reinforcement learning. While tutorials and courses are helpful, reading the original papers can also help you become acquainted with the original ideas and caveats that come with some of these models and approaches. We found that OpenAI has a “Key Papers in Deep RL” site that has tons of great papers about RL algorithms.
Fun Video - Simulating Natural Selection
The concept of evolutionary algorithms spiked my interested in bots quite a few years ago. Although the focus in the community has recently been centered around LLMs and RL approaches, it’s really awesome revisiting these videos and drawing inspiration from them. Check out this video that simulates natural selection for a good primer on these approaches.
Technical Paper of the Week
A beautiful aspect of the AI + games field is that it advances our progress in real-life use cases that can potentially be solved by bots. An example of this is self-driving cars - after all, a car is just an agent behaving in some environment, although with way more risk! This paper from 2020 summarizes some of the work done to apply RL algorithms to self-driving cars, called “Deep Reinforcement Learning for Autonomous Driving: A Survey”.
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.