I'm a CS student at UC Berkeley who's currently taking a gap away from school to self-learn with my own curriculmn. This blog is to document and organize my learning.
I was born and raised in Seoul, Korea up until middle school, but went to high school in Jeju island-a volcanic island at the southern tip of Korea. I spent a lot of time biking and diving in the waters. Seeing the beauties of the waters made me want to study marine science or physics in college.
A lot changed my first semester there. I took an introductory cs class taught in Snap!(berkeley's version of scratch) with the nudge of a friend and loved it. Maybe it was the combination of having a friend to do fun projects with along with many dorm friends also taking cs classes.
When covid struck I came back to Korea and served in the Korean Navy. I was on a battleship as a boatswain's mate, but towards the end of my service I met a friend who introduced me to ML. It was a happy combination of science and engineering. After my service, I did my first ML project in underwater CV where I turned my distorted diving photos into ones that were clean.
I spent most of my time reading papers instead of classwork. I wandered through lots of rabbit holes ranging from monocular depth estimation, multi-agent RL, diffusion, hyperbolical NNs, and GNNs. They were mostly methodology papers, but I realized there's a lot of overlooked engineering that's not specified well in these papers.
I'm taking a gap back in seoul studying the engineering behind ML, especially memory-optimization and efficient scaling methods.
"In the novelist’s profession, as far as I’m concerned, there’s no such thing as winning or losing…
What’s crucial is whether your writing attains the standards you’ve set for yourself.
Failure to reach that bar is not something you can easily explain away.
When it comes to other people, you can always come up with a reasonable explanation, but you can’t fool yourself.”
- from haruki murakami's What I Talk About When I Talk About Running
If any of this interests you, please reach out!
email: ko.hyeonmok at berkeley dot edu
github: henryhmko
The formats in this blog were inspired by Lilian Weng, Simon Boehm, and Fabien Sanglard.