Picture of my computer with the HTML code for my distill.pub publication. Background is a window showing Tokyo, Japan.
Picture of my computer with the HTML code for my distill.pub publication. Background is a window showing Tokyo, Japan.
November 2018. Finishing my first publication, “Visualizing memorization in RNNs,” published in distill.pub. This was my first publication. I thought that and a Master's degree in Machine Learning would be enough to get a Research Software Engineer position; I was very wrong! — Tokyo, Japan

Becoming an Independent Researcher and getting published in ICLR with spotlight

Why I became an Independent Researcher, why you should avoid it, and advice for those who make the difficult choice anyway.

Tweeting my ICLR publication. After that, I got almost 100 messages from Independent Researchers asking for advice.

Why I became an Independent Researcher

Reality check, “1–2 publications in top ML venues”

To start a PhD, without insider referral, you need to do work equivariant to half of a PhD.

Getting started, how I funded myself

To say I lucked out regarding funding, would be an understatement. But Denmark is also an incredibly expensive country. Funding myself in another country would probably have been easier.

Getting started, where do ideas come from

Almost all publications exaggerate how well they perform. Just improving others work is a viable research strategy.

A lonely life, not losing my sanity or hope

Not having a support network that are experiencing the same struggles of writing a paper, as first-author, is the number one reason I would recomend against becoming an Indepedent Researcher.

Picture of my hand on a huge bamboo, not being able to cover (read, handle) the bamboo at all.
Picture of my hand on a huge bamboo, not being able to cover (read, handle) the bamboo at all.
Publishing as an Independent Researcher was probably more than I could handle (literary pun intended). I expected failure, but tried anyway and gave it my best. — Somewhere, Japan
Example of a 3-day side-project. One person reached out and was thankful; just that is a valuable source of encouragement.

Writing a great paper

There will be 1–2 messages in a paper, that if misunderstood, will completely confuse the reader and be the first cause of getting rejected. Do not be afraid of repeating a message to prevent that.

Rejected from NeurIPS 2019

Having anonymous peer-reviewers hold your life in their hands is so strange.

“Red ink” by Jorge Cham www.phdcomics.com
“Red ink” by Jorge Cham www.phdcomics.com
Me reading reviewer #1 (yeah, #1 and #2 swapped places) comments from NeurIPS 2019.
“Red Ink” by Jorge Cham — www.phdcomics.com

Reviewers are likely to side with already published results. They will only think critical about your submission not previous publications, especially if they come from DeepMind.

Submitting to ICLR, what was different

Worth the effort?

Long hanging bridge, where the end can not be seen.
Long hanging bridge, where the end can not be seen.
The path to Machine Learning Research is long for all. However, when you can’t even get off the ground, it can seem impossible. If you really want to, spend the effort and look for alternative ways. — Vancouver, Canada

Q & A

Summarizing

Image for post
Image for post
Working as an Independent Researcher is incredibly hard and quite lonely. However, with luck, there might be light around the corner. — Vancouver, Canada

For the researchers, who may also be reading along

If you finished your Master’s in 2017 or earlier, getting into a good PhD program was achievable. Now it requires 1–2 publications in top venues (NeurIPS/ICLR/ICML), preferably with famous co-authors. I hope researchers, professors, committee members, and conference organizers, will help to stop this new elitism that is rapidly developing! It only amplify the existing biases that are already challenging the industry.

9 months after the publication, what happened

Written by

Ph.D. Student at Mila, researching interpretability for Machine Learning because society needs it. Published in Distill, SEDL|NeurIPS, and ICLR (spotlight).

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store