For this first challenge, we use Unity MLAgents 2vs2 Soccer ⚽, an environment made by the MLAgents team.
We want to organize more multi-agent challenges in the future, so don't hesitate to give us feedback 🤗
Participate in the challenge here 👉 huggingface.co/deep-rl-course
3/3
Follow
Thomas Simonini
@ThomasSimonini
Developer Advocate 
|
Love creating AI in games using NLP and RL 
|
Founded Deep Reinforcement Learning Course 
bit.ly/3FJEPgI
|
Love creating AI in games using NLP and RL |
Founded Deep Reinforcement Learning Course bit.ly/3FJEPgIThomas Simonini’s Tweets
With our new AI vs. AI tool made by , when you push your trained model to the Hub, it gets evaluated and ranked against others.
🏆 Matchmaking algorithm to define matches and leaderboard
👀Visualize your agent competing with our demo 👉 huggingface.co/spaces/unity/S
2/3
1
6
Show this thread
Introducing ⚔️ AI vs. AI ⚔️ a deep reinforcement learning multi-agents competition
⚽️ Learn about Multi-Agents and train your team to play Soccer.
🏆 Compete against other classmates' teams.
Participate in the challenge here 👉 huggingface.co/deep-rl-course
1/3
6
76
372
Show this thread
We will be hosting a book reading club this Saturday, on the classic Reinforcement Learning book chapter 1 & 2 by Sutton and Barto, join our discord here 👉🏼 hf.co/join/discord and find it on the events tab! 📓🧑🏼🏫
4
12
71
First Challenge: Unity MLAgents 2vs2 Soccer ⚽️, an environment made by the Unity team.
Starting this Wednesday in the Deep RL Course 👉huggingface.co/deep-rl-course
2/2
1
8
75
Show this thread
At , we're preparing something big 🤯
⚔️AI vs. AI ⚔️
A multi-agent challenge where your deep reinforcement learning agents compete against one another
🏆 Matchmaking algorithm to define matches, and leaderboard
👀Visualize your agent competing with our demo
1/2
21
163
1,059
Show this thread
Image2Image makes Stable Diffusion a powerful tool for illustrators. Here's how to incorporate it in your 2D asset workflow.
1
20
57
🎮NEW VIDEO🎥
For the 70th episode of 'AI and Games', it's time to return to Naughty Dog's post-apocalyptic franchise and dig deep into how AI is used to tell the story of #TheLastOfUs Part Two.
Watch on YouTube:
youtube.com/watch?v=BghECm
#gamedev #gaming #YouTube
13
21
The new era of all things AI and Games kicks off this week, including:
- AI and Games #70 analysing The Last of Us Part II
- Branching Factor #1 on AI and Games Plus and all major podcast providers.
- And Smoke & Mirrors too!
Subscribe, sign up and get ready for more STUFF!
1
3
10
Show this thread
Happy to share my paper on Learning to Exploit Elastic Actuators for Quadruped Locomotion.
We learn to trot/pronk in only 10 minutes, directly on the real robot 🐈.
Paper: arxiv.org/abs/2209.07171
6
18
115
Show this thread
Excited that my hobby solo-project city builder will be in Steam Base Builder Fest
1
5
Hello 👋🏼
I just uploaded all of my cheatsheets (there's 100+ of them!) on various topics including machine learning, data structures, statistics and more.
📔 GitHub repository: github.com/merveenoyan/my
📓 Hugging Face dataset repository: huggingface.co/datasets/merve
20
159
910
Show this thread
3D Asset Generation: Learn about where we are now, and where it's going:
3
41
140
// "Deep Papers" Podcast + Next Paper? //
Introducing the "Deep Papers" podcast series: deep dives into AI papers with AI scientists.
First episode, out tomorrow: We interview and , scientists at OpenAI behind InstructGPT: the precursor to ChatGPT!
1/2
5
19
149
Show this thread
The sixth Unit of the Deep Reinforcement Learning course has been published 🥳
You’ll learn about Advantage Actor Critic (A2C) and you’ll train robots 🤖 to walk and reach objects with their arms
Sounds fun? Start learning now 👉 huggingface.co/deep-rl-course
2
37
181
Show this thread
Stable-Baselines3 v1.7.0 is out =)
It comes with better handling of 3D input, many bug fixes and quality of life improvements.
We are also currently running a complete benchmark of all algorithms on all env for the RL Zoo, see below for more information.
1
5
26
Show this thread
The fifth Unit of the Deep Reinforcement Learning course has been published 🥳
You’ll learn:
✅ How ML-Agents work.
✅ Train Julien the 🐻 to shoot snowballs to target
Sounds fun? Start learning now 👉 huggingface.co/deep-rl-course
3
50
249
If you have questions, please ask below I'll reply 🤗 ⬇️
Keep Learning and stay Awesome 😎
4
6
Show this thread
And if you want to join a community of reinforcement learning enthusiasts and researchers. Join our discord. The RL team will be there 👉 hf.co/join/discord
1
1
7
Show this thread
So now is the perfect time to start 🥳
We wrote an introduction unit to help you get started 👉
1
13
Show this thread
Finally, in an optional unit, you'll learn about advanced topics such as:
✔️ Decision Transformers
✔️ Offline RL
✔️ Model-based RL
✔️ MineRL environment
✔️ RLHF the technology behind ChatGPT
Quote Tweet
Interested in RLHF? So are we!
We're adding some RLHF content to the @huggingface Deep Reinforcement Learning Course.
Join us live next Tuesday to learn more -- RLHF: From Zero to ChatGPT.
When: Next Tuesday 13 Dec, at 5:30pm CET / 11:30 am ET
Where: youtu.be/2MBJOuVq380
When: Next Tuesday 13 Dec, at 5:30pm CET / 11:30 am ET
Where: youtu.be/2MBJOuVq380
1
1
6
Show this thread
In Unit 8, you'll learn about PPO and train agents playing VizDoom (open source Doom FPS) and environments we created
Quote Tweet
The last environment for my @godotengine Reinforcement Learning lib, a team FPS.
The release is planned for tomorrow once I have updated the docs.
github.com/edbeeching/god
Env source code and builds are already available on the @huggingface hub:
huggingface.co/datasets?sort=
github.com/edbeeching/god
Env source code and builds are already available on the @huggingface hub:
huggingface.co/datasets?sort=
0:20
4.5K views
1
1
6
Show this thread
In Unit 7, you'll learn about Multi-Agents and participate in the first AI vs AI Challenge 🏆 where you'll train your agent to learn to play snowball fight ☃️
But also a soccer game ⚽
GIF
1
7
Show this thread
In Unit 6, you'll learn about A2C and train your agent using Stable-Baselines3 in robotic environments.
✔️ A bipedal walker 🚶 to learn to walk.
✔️A spider 🕷️ to learn to move.
✔️A robotic arm 🦾 to move objects in the correct position.
1
6
Show this thread
In Unit 5, you're going to learn about Unity MLAgents and train two agents.
1
4
Show this thread
In Unit 4, you'll study policy gradient, and code it from scratch using PyTorch 🔥
You’ll then train this agent to play PixelCopter 🚁 and CartPole
The Unit 👉 huggingface.co/deep-rl-course
1
3
Show this thread
In Unit 3 and Bonus Unit 2 you'll learn about Deep Q-Learning and train your agent to play Atari games using RL-Baselines3-Zoo 👾 and Optuna, a hyperparameter search library.
The Unit 👉 huggingface.co/deep-rl-course
1
1
Show this thread
In Unit 2 you'll study value-based methods, Q-Learning and build an Autonomous Taxi 🚖
The Unit 👉 huggingface.co/deep-rl-course
1
4
Show this thread
In Unit 1 and Bonus 1 you:
✔️ Study the foundations of Reinforcement Learning
✔️ Train a Lunar Lander agent to land on the moon 🌕
✔️ Train Huggy the 🐶 to fetch a stick and play with him directly in your browser 💗
Start Learning Now 👉 huggingface.co/deep-rl-course
1
1
6
Show this thread
And to start learning about Deep Reinforcement Learning, we wrote a free course.
📖 🤖 A series of 8 units with theory part and hands-on.
👩🎓 Certificate of completion.
⚔️ AI vs AI challenges where you'll train your agents to compete against other's classmate's agents.
1
1
9
Show this thread
Deep RL is the most fascinating topic in AI and it's getting a lot of attention recently with ChatGPT.
Indeed, ChatGPT uses reinforcement learning (RLHF) in his training. And thanks to this technique it got very good results.
So RL is an important skill to master this year
Quote Tweet
Interested in RLHF? So are we!
We're adding some RLHF content to the @huggingface Deep Reinforcement Learning Course.
Join us live next Tuesday to learn more -- RLHF: From Zero to ChatGPT.
When: Next Tuesday 13 Dec, at 5:30pm CET / 11:30 am ET
Where: youtu.be/2MBJOuVq380
When: Next Tuesday 13 Dec, at 5:30pm CET / 11:30 am ET
Where: youtu.be/2MBJOuVq3801
2
8
Show this thread
Why do you need to start studying Deep Reinforcement Learning now?
The free Deep RL Course: huggingface.co/deep-rl-course
A thread 🧵👇
5
124
535
Show this thread
We just released the world's first open-source humanoid physical strike AI! The AI is trained with large-scale self-play RL with a pre-trained ASE model from . Thank for this amazing work. github.com/inspirai/TimeC
10
91
472
What is Wave Function Collapse? It's a really cool algorithm that generates stuff. I made this video on it a few months ago
1
3
10
The 4th Unit of the Deep Reinforcement Learning course has been published 🥳
You’ll study policy gradient, and code it from scratch using PyTorch 🔥
You’ll then train this agent to play PixelCopter 🚁 and CartPole
Start learning now 👉 huggingface.co/deep-rl-course
2
37
168
There were much more events and projects during 2022 but I'll stop there 😸 .
If you want to join a community of reinforcement learning enthusiasts and researchers. Join our discord. The RL team will be there 👉 hf.co/join/discord
1
1
6
Show this thread
Also in December host a live and wrote some blogposts about RLHF, the technology behind ChatGPT.
Quote Tweet
Interested in RLHF? So are we!
We're adding some RLHF content to the @huggingface Deep Reinforcement Learning Course.
Join us live next Tuesday to learn more -- RLHF: From Zero to ChatGPT.
When: Next Tuesday 13 Dec, at 5:30pm CET / 11:30 am ET
Where: youtu.be/2MBJOuVq380
When: Next Tuesday 13 Dec, at 5:30pm CET / 11:30 am ET
Where: youtu.be/2MBJOuVq3801
1
5
Show this thread
In December, we launched a new version of the Deep Reinforcement course with new environments, new concepts (such as multi-agents), and AI vs AI challenges ⚔️
It's not too late to join the course 👉 huggingface.co/deep-rl-course
Quote Tweet
We're launching a new version of the @huggingface Deep Reinforcement Learning Class 
From beginner to expert
New libs: Sample Factory and CleanRL
New environments
AI vs Humans contests
And more!
Starting December the 5th
Sign up here eepurl.com/ic5ZUD
From beginner to expert
New libs: Sample Factory and CleanRL
New environments
AI vs Humans contests
And more!
Starting December the 5th
Sign up here eepurl.com/ic5ZUD
2
2
8
Show this thread