Source of Interest Group: Reinforcement Learning


  Status: **active**

We currently have 3 channels on Discord: 

### Channel [`#reinforcement-learning`]() <br>
Channel for general RL discussions, organisation of paper readings and other RL events. Check the MLC calendar for details of upcoming events. <br>
#### Papers
| **Date**       | Description | 
| ----------- | ----------- | 
| Jun - Aug 2022 | We are reading another long format paper: "Emergent Bartering Behaviour in Multi-Agent Reinforcement Learning" over several sessions  |
| Dec 2021-Jan 2022 | We completed "Acquisition of Chess Knowledge in AlphaZero" over several sessions  |
| Papers in 2021   | We covered some core RL papers including [PG](https://docs.google.com/presentation/d/1WpGhIJ5Kf1puWwlwaLcLyY4ymSlg1KrOEy7m4kyjYL8/edit#slide=id.ge4e1b7dc24_0_3), [DQN](https://docs.google.com/presentation/d/16GUfTvMGD2QPTVDzAMNlxZKFrDw1ib3yGsybuDwyw54/edit#slide=id.p), [PPO](https://docs.google.com/presentation/d/1SyvJ-XJUbMboKup8xHXOdSHDPXLW95_NIMEMLR7ghhc/edit#slide=id.ge68055b75a_0_3) |

### Channel [`#rl-implementation`]() <br>
Channel for people who want to implement RL algorithms and work on code. This is a practical group so be prepared to code, learn and teach! <br>

We meet 3 times a week. Check the MLC calendar for details. <br>

#### Projects:
| **Date**       | Description | 
| ----------- | ----------- |  
| In Prog Dec-Jan 2022 | Implementation of PPO from scratch. See git [repo](https://github.com/ML-Collective/rl-playground/tree/main/ppo) |
| Oct/Nov 2021 | Implemented A3C as coding groups  |

<br>
After working on some core algorithms we will tackle papers more generally and challenges normally hosted by conferences. 
<br>

### Channel [`#rl-beginners`](): <br>
This channel is basically for the people who just started their journey in RL. Initially, we are trying to design our own curriculum that we'll follow and plan further once we are done with beginner-level studies. <br>
1. Building a foundation on RL by reading/implementing together <br>
2. Reading papers (beginner level) <br>
3. Implementing papers <br>
4. Working on project/research collaboratively. <br>
We have a week-by-week plan to move forward and will share it shortly.