I would! If you are thinking to write one, do it!

But... they are equivalent?

Yeah I know, but I thought I was doing purely theoretical excercises.

And we always changed the regex NFA to an equivalent DFA and that was the implementation.

So somehow I managed to internalize the idea that an NFA is purely theoretical and can't be built.

Modulo an exponential blowup! That’s like saying P is equivalent to NP.

Depends on what you mean by that. You can convert every NFA into a DFA. That's a NP complete (IIRC), but running the DFA is O(n). Running the NFA without converting it is also NP complete. One isn't better than the other, but the costs vary for different expressions and usages.

Running NFA is O(nm) not NP.

Sorry, you're right. Capturing worst case was much more expensive, I believe, but I'm no longer sure.

So it is NP (in fact P)

The blow up is exponential for carefully crafted academical regular expressions.

im practice is a good idea to build a DFA from your regex, up front (re2) or lazily (ripgrep)

No, because you can compute the optimal automaton (as in least number of states) that recognizes the same language: https://en.wikipedia.org/wiki/DFA_minimization

Was it a controlled study or just correlation?

Huh? Triton inference server and Triton the kernel language are two distinct, very different things… Is this AI-generated?

Just yesterday I was reading through this five year old post on triton by its creator. Triton was their PHD thesis and they coined the name before the inference server was renamed to it… if this is what you are referring to?

Here is the Reddit thread:

https://www.reddit.com/r/MachineLearning/comments/otdpkx/n_i...

> What actually happens: random Internet users spend two seconds skimming, then click their favorite.

> They're not reading carefully. They're not fact-checking, or even trying.

Uhhh, how was that established?

cuz area and power

Area and power are why there was a choice to make. AI data centre demand is why they made this choice specifically.

There's antimony, arsenic, aluminum, selenium…

What's the point of the relu in the loss function? Its inputs are nonnegative anyway.

Let's try to keep things positive.

I wondered the same. Seems like it would just make a V-shaped loss around the zero, but abs has that property already!

RELU would have made it flat below zero ( _/ not \/). Adding the abs first just makes RELU do nothing.

In reality it’s probably not a RELU modern LLMs use GeLU or something more advanced.

Indeed, we should call all idempotent functions twice just in case the first incantation fails to succeed.

In all seriousness, this is not at all how resilience to cosmic interference works in practice, and the probability of any executed instruction or even any other bit being flipped is far greater than the one specific bit you are addressing.

Well, I guess it's helping to distinguish authors who are doing arithmetic they understand from ones who are copying received incantations around...

"delves"

dashes

an explicit "conclusion" section at the end

Humans do not write conclusions? As someone who went to college, that is a natural way to end a long essay. True mark of higher education would be writing the conclusion at the top.

Exactly, it's a natural way to write a college essay. I've never not cringed reading an article/blog post that is structured that way, it comes across very contrived. I've also noticed that LLMs tend to prefer it, and humans tend to avoid it in general.

For a 500 word blog post, sure it may be a bit much. This article was a decently hefty read, for which the summary callouts are not out of place.

It's called an executive summary in that case, ain't it?

I was thinking an academic Abstract, but sure.

> dashes

I am forever angry that LLMs have ruined em-dashes. They’re a wonderful part of punctuation.

None of which are hard AI tells.

Lisp In Small Pieces

Yep this is the best way into the heart of LISP

Seconded, it's a great book.

Thirded.