Implement Papers

One of thing I want to achieve from part2 of the course is to implement paper. Is anyone interested in the idea of collaborating and implementing papers? We can create virtual group and share ideas.


Excellent idea!

It is a good idea! I would be interested to participate!

definitely interested about learning more about how great and useful papers are produced :wink:

Yes, totally!

Same! When I learned about research papers during the course, I never thought I’d be able to understand methods used. After finishing the course, I understand them now, but I think I’ll need a team to implement them. Now here’s the team!

Can’t wait to team up and contribute!

on a tangent, it would be great to have a thread/place to share some passage in the paper you’re puzzled/not sure how to interpret/implement and ask for advice…


another thought - most of top tier papers come with implementations…
i was toying with the idea of reverse engineering those back to the paper as the easiest way to gain the understanding, the math part esp…


Can anyone help us create a slack group or something so that we can meet online once in a week? Please let us know what date or time is good to discuss and share ideas?

Here is the slack channel I created:

1 Like

I was wondering if we work together to implement a few paper then we will get confidence to start working independently.

BTW Helena your artistic work is great!

1 Like

We already have a small TWiMLAI community, I’d suggest that we could discuss the same there :slight_smile:

1 Like

It would be a great way to dive deeper into the nitty-gritty of papers

Sure we can use TWiMLAI community. Thanks

I would propose translating a paper into a Colab notebook +/- using Github, so that there is a tangible product and goal for the group. Would benefit the group and those coming after you.

what is TWiMLAI and how can I join ? =)

This Week in Machine Learning & AI

1 Like

You can join here



would be a good reference.

This looks very good. Should we take papers where code is written in Pytorch and meet one day in a week to discuss the paper?