We’re excited to share our first custom Deep Reinforcement Learning environment: Snowball Fight 1vs1 🎉.
Snowball Fight is a game made with Unity ML-Agents, where you shoot snowballs against a Deep Reinforcement Learning agent. The sport is hosted on Hugging Face Spaces.
👉 You’ll be able to play it online here
On this post, we’ll cover the ecosystem we’re working on for Deep Reinforcement Learning researchers and enthusiasts that use Unity ML-Agents.
Unity ML-Agents at Hugging Face
The Unity Machine Learning Agents Toolkit is an open source library that permits you to construct games and simulations with Unity game engine to function environments for training intelligent agents.
With this primary step, our goal is to construct an ecosystem on Hugging Face for Deep Reinforcement Learning researchers and enthusiasts that uses ML-Agents, with three features.
- Constructing and sharing custom environments. We’re developing and sharing exciting environments to experiment with recent problems: snowball fights, racing, puzzles… All of them can be open source and hosted on the Hugging Face’s Hub.
- Allowing you to simply host your environments, save models and share them on the Hugging Face Hub. We have now already published the Snowball Fight training environment here, but there can be more to return!
- You’ll be able to now easily host your demos on Spaces and showcase your results quickly with the remaining of the ecosystem.
Be a part of the conversation: join our discord server!
In the event you’re using ML-Agents or fascinated about Deep Reinforcement Learning and wish to be a part of the conversation, you possibly can join our discord server. We just added two channels (and we’ll add more in the longer term):
- Deep Reinforcement Learning
- ML-Agents
Our discord is the place where you possibly can exchange about Hugging Face, NLP, Deep RL, and more! It is also on this discord that we’ll announce all our recent environments and features in the longer term.
What’s next?
In the approaching weeks and months, we can be extending the ecosystem by:
- And we’re constructing recent custom environments that can be hosted in Hugging Face.
Conclusion
We’re excited to see what you are working on with ML-Agents and the way we will construct features and tools that aid you to empower your work.
Do not forget to join our discord server to be alerted of the brand new features.
