Thread by @gordic_aleksa

I just open-sourced my implementation of the original @DeepMind's DQN paper! But this time it's a bit different!

There are 2 reasons for this, see the thread.

GitHub: github.com/gordicaleksa/p…

#rl #deeplearning

1) This time the project is still not completely ready**. I'm yet to achieve the published results - so I encourage you to contribute!

Many of you have been asking me whether you can work on a project with me and I'll finally start doing it that way - from now onwards. ❤

2) This repo has the ambition to grow and become the go-to resource for learning RL. So collaborators are definitely welcome as I won't always have the time myself.

** main reasons are:
a) I was very busy over the last 2 weeks
b) It currently takes ~5 days to fully train DQN

for some reason I get serious slowdowns - checkout the README I've described everything you need to know.

As always I really hope this project will resonate with the community and that people find it useful!

With this project, I've completed every single deep learning project I had on my list since March 2020.

6 projects in total were on that list. But since I've made 3 NST projects I ended up open-sourcing 8 projects haha.

1) NST (original, Johnson, videos)
2) DeepDream
3) GANs (original, cGAN, DCGAN)
4) Original Transformer (Vaswani et al.)
5) Graph Attention Network (Velickovic et al.)
6) Deep Q Network (Mnih et al.)

Extremely productive 14 months! 😅

Keeping in mind that that's definitely not everything I do I'm very happy with what I've done and with the way the community accepted these projects.

Hearing from so many of you how these helped you on your own journey feels really great and I'm humbled and grateful for that.