Are you interested in learning more about reinforcement learning in artificial intelligence, but you are not sure where to start? If this sounds like something that interests you, then this hackathon event is for you!
Reinforcement Learning Hackathon by Deep Learning Labs
This is a beginner-friendly event that will teach you the basics of reinforcement learning using stable-baselines, Google colab & OpenAI Gym.
Hackathon Details
This is a one-day hackathon. It takes place online on the 23rd of July, Saturday. The event starts at 10:00 AM CEST with an introduction to Reinforcement Learning and ends with demo presentations at 5:00 PM CEST.
Technology
Reinforcement learning is a subfield of machine learning that focuses on training agents to make optimal decisions in an environment by maximizing some reward function.
We will teach you how to use RL in theory and then you will put your knowledge into practice. We will share with you some boilerplates that will assist you in getting started. Furthermore, our experts will be available for you throughout the entire event.
Participants
This hackathon is for beginners in AI, but some coding experience is recommended. However, don't worry if you're not an expert coder, as there will be plenty of help available throughout the event.
How to participate in the hackathon
The hackathon will take place online on the lablab.ai platform and Deep Learning Labs Discord Server. Please register for both in order to participate. To participate click the "Enroll" button at the bottom of the page.
Challanges and Prizes
As with every hackathon we will have a number of challenges and prizes. The prizes will be awarded to the top performers after given criterias.
Challange scoring
We will score the hackathon after following criterias.
There is 3 enivorments
- Lunar Lander
- Cartpole
- Bipedal Walker
The 3 reinforcment learning challenges in the hackathon that are scored as follows:
-
Top 3 results in each challenge will receive a prize of 1-10 points whereas the best result will retrive 10 points.
-
Additionally we will give +3 points for each one of the challanges if you manage to record a video of your agent playing the game.
-
To qualify for the prize you need to be able to explain how you achived the results of your agent during Q&A with judges
Resources
Here is a couplple of resources that will help you get started with the hackathon. We deeply recommend that you check them out
Colab Use colab to run your code.
stable-baselines stable-baselines is a tensoflow library for reinforcement learning. stable-baselines3 stable-baselines3 is a pytorch library for reinforcement learning.
OpenAI Gym OpenAI Gym is a library for reinforcement learning.
Deep Learning Labs Discord Server The Deep Learning Labs Discord Server.
gymlibrary a up-to-date Gym library
Hackathon prizes
A total of $2,000 in Digital Ocean credits to the winning teams.
Video from our previous event
About Deep Learning Labs
Deep Learning Labs, is a popular series of on-site and online hackathon events. Our mission is to inspire hackers to build new solutions based on the most powerful tech on the planet.
Hackathon FAQ
Who can join the Hackathon?
We welcome domain experts from all industries, not just AI/tech. Successful AI solutions require a combination of technical expertise and domain knowledge. Coding experience is recommended.
Do I need a team?
You are welcome to join as a team or solo, if solo. We encourage you to look for a team before the event. We recommend you to join the Deep Learning Labs Discord channel: https://discord.gg/gCuBwBB35k and posting in the #looking-for-team channel to get to know your potential future team members
Do I need a Github account?
It is recommended, that at least one team member has a Github account. You can create one for free if you don't already have one.
I have other questions.
Feel free to reach us on social media, @deeplearnninglabs, or through our Discord channel.
Event Schedule
- To be announced
Submitted concepts, prototypes and pitches
Submissions from the teams participating in the Reinforcement Learning OpenAI Gym event and making it to the end 👊
.png&w=3840&q=75)
.png&w=3840&q=75)
Cartpole, LunarLander, BiPedal Walker
Applied reinforcement learning for agent to play these 3 games: Cartpole, LunarLander, BiPedal Walker. We used the basic model of Environment -> State -> Agent -> Action to train our agent. We reward the agent for achieving an outcome that we want, while penalizing the agent for doing otherwise. After many iterations, our agents learns to clear the games.
Voyager
.png&w=3840&q=75)
Reinforcement Learning: CartPole, Lunar Lander and Bipedal Walker
Using the stable_baselines3 library, we tried to solve the problems proposed in the challenge. We used a Proximal Policy Optimization (PPO) Model. The Policy we used is a standard MLP. We tried to change the number of iteration to achieve a better performance.
Ack-Ack Learn
.png&w=3840&q=75)
Mesa-optimizer team's submission to the RL hackathon
Used A2C and DQN for Lunar Lander DQN for Cartpole TQC for Bipedal Walker
Mesa-optimizers
.png&w=3840&q=75)
Los patxatxos
.png&w=3840&q=75)
Radio Frequency Project
We have completed 2 challenges. The first one (cartpole) was completed using our own code, we implemented Deep Q Learning. For the second one (Lunar Lander) we used stable_baseline library.
RadioFrequency
Teams: Reinforcement Learning OpenAI Gym
Check out the roster and find teams to join
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
