Silamoth

I recently read "Enabling tabular deep learning when d ≫ n with an auxiliary knowledge graph" (https://arxiv.org/pdf/2306.04766.pdf) for one of my graduate classes. Essentially, when there are significantly more data points than features (n >> d), machine learning usually works fine (assuming data quality, an underlying relationship, etc.). But, for sparse datasets where there are fewer data points than features (d >> n), most machine learning methods fail. There's just not enough data to learn all the relationships. This paper builds a knowledge graph based on relationships and other pre-existing knowledge of data features to improve model performance in this case. It's really interesting - I hadn't realized there were ways to get better performance in this case.

My thoughts exactly. The author seems to think a proprietary period is the best option. But honestly, I'd rather have data be immediately openly available and ensure we credit those who played an important role in identifying what observations should be made and obtaining the data. That really seems like the best of both worlds. Promoting open access to data and collaboration among scientists is better than promoting closed access to data and "scooping" of results/limited collaboration.

Writing blog posts has been in my back-of-the-mind TO-DO list for a while now. I just never seem to have time for it. Or rather, there are a million things I end up prioritizing over blogging. One of these days, when I'm less busy, I'll start writing some blog posts. I definitely have some ideas for posts that I think could be interesting.

But reading papers is fun and cool. You get to learn new things. If you read recent papers, you learn things that aren't just new to you but new to the entire world; you can literally learn cutting-edge things that were discovered recently. And if you read older papers, you can better understand why the world is the way it is right now.

Putting TC aside (don't most programmers make plenty anyways?), isn't it just cool to learn new and cutting-edge things? Personally, I think so. Curiosity is central to what it means to be a hacker.

Very interesting paper! It's cool to see that Fortran is still evolving. I've been considering learning it (when I have the time) because of its use for numerical computations and parallel processing. I feel like there's been a big push towards general-purpose languages like C++. However, I'd love to get a better feel for what it's like to program in a language specifically designed for your use case. The closest I've experienced is MatLab, but I feel like MatLab is designed more for scientists and engineers who need to perform some calculations but don't want to actually learn how to program, not for professional programmers working on scientific computation.

Honestly, I don't know why Microsoft thought this was a good idea in the first place. Casual users don't know the difference between browsers and will just use the default anyways. But power users are going to use our preferred browser one way or the other. If it's difficult to set the default browser on Windows 11, then that's just one more reason we'll move to a better operating system.

That was a great article! Pipes can definitely be very powerful. I will say, though, that I often find myself reading pages of documentation in order to actually get anything with Unix and its many commands.

Thanks for sharing! I'll have to give that a read sometime. It looks quite interesting.