Hacker News: traes

New comment by traes in ""Drawing" the Mona Lisa with GPT-5.6, Claude, Gemini, and Grok"

traes — Wed, 22 Jul 2026 08:59:38 +0000

Perhaps there was something in the prompt or settings preventing this, but I'm surprised (and slightly disappointed) that none of the models approached this the way I would: download an image of the Mona Lisa, run all of the drawing functions many times to construct a forward model of the drawing implements, and attempt to solve some kind of explicit inverse problem through either ML or a classical algorithm to minimize some difference metric. Were they just restricted from running code or are the models unindustrious without a very specific prompt?

New comment by traes in "GPT-5.6 used a prompt to close a 30-year gap in convex optimization"

traes — Sun, 19 Jul 2026 11:32:14 +0000

That is exclusive to engineering PhDs, entirely irrelevant to the discussion of math PhDs.

New comment by traes in "Completing a computer science degree on Coursera"

traes — Mon, 06 Jul 2026 01:03:37 +0000

As someone who has met a lot of math majors and a lot of CS majors, I am skeptical of your supposition that CS majors are better at finding and applying novel abstractions than math majors who know how to code.

New comment by traes in "International chess federation sanctions Kramnik"

traes — Sat, 04 Jul 2026 09:05:48 +0000

I didn't watch it myself, but there was a recent Netflix documentary on the subject where Magnus claims he was misled by chess.com into believing that they had more concrete evidence than they did.

New comment by traes in "Odin, Wikipedia and engagement farming"

traes — Sat, 04 Jul 2026 07:00:37 +0000

From Cambridge dictionary:

obscure (adjective)

not known to many people:

- an obscure island in the Pacific

- an obscure 12th-century mystic

Why does its use in production matter? Perhaps the syntax itself is obscure, but we're not discussing syntax but general awareness. Anyway, the most common "real" use of brainfuck is to prove Turing completeness of other things by finding a way to compile them into brainfuck.

New comment by traes in "Odin, Wikipedia and engagement farming"

traes — Sat, 04 Jul 2026 06:55:55 +0000

What? Brainfuck is the single least obscure esoteric programming language. It's the most famous example of a simple Turing complete language and its provocative name gets it a fair amount of media coverage outside its niche.

New comment by traes in "International chess federation sanctions Kramnik"

traes — Sat, 04 Jul 2026 03:38:41 +0000

This is true, but the crushing response to this scandal is clearly disproportionate. Millions of views on videos making crude sexual jokes about a 19 year old is already an incredibly unfortunate situation, and his career got completely halted for years due to (what are now clearly) false accusations.

New comment by traes in "International chess federation sanctions Kramnik"

traes — Sat, 04 Jul 2026 03:36:08 +0000

> he probably did cheat in this particular game

I have no idea why this continues to be a popular opinion among laypeople. Even at the time actual cheating experts using computer analysis determined it to be likely fair play. His accuracy was rather pedestrian, Magnus just had on off day and played terribly. There is no proposed mechanism for how he could even HAVE cheated. He has continued to perform at a similar level under much tighter security enforcement. The primary accusation seems to be that he's poor at explaining the reasoning behind his moves, but anyone who's ever watched a Hans stream can tell you that the man simply cannot speak.

New comment by traes in "International chess federation sanctions Kramnik"

traes — Sat, 04 Jul 2026 03:31:25 +0000

Here's a fun compilation of him getting cooked by 12 year olds and then immediately reporting them:

https://www.youtube.com/watch?v=Gg5LSOoYoPs

New comment by traes in "International chess federation sanctions Kramnik"

traes — Sat, 04 Jul 2026 03:25:55 +0000

Regarding general knowledge, it's a pretty well known phenomena within the community. Here's a video with the three players who spent the longest time at #2 in the world (see [0]) behind Magnus Carlsen answering basic trivia questions[1].

[0]: https://www.reddit.com/r/chess/comments/1uk3wa3/world_no_2s_...

[1]: https://www.youtube.com/watch?v=sXyLJaaLEpE

New comment by traes in "International chess federation sanctions Kramnik"

traes — Sat, 04 Jul 2026 03:20:34 +0000

A lot of players actually do go to college, although likely a minority. I know that Maxime Vachier-Lagrave in particular has a math degree. Ding Liren has a law degree. Daniel Naroditsky himself, while not quite a top grandmaster by the typical definition, had a bachelors in history.

New comment by traes in "Deriving the SVD (Single Value Decomposition) from scratch"

traes — Wed, 01 Jul 2026 05:41:38 +0000

So at least the user who submitted it to HN was a human (or perhaps a very scheming LLM attempting to appear it?)

New comment by traes in "Deriving the SVD (Single Value Decomposition) from scratch"

traes — Wed, 01 Jul 2026 05:40:30 +0000

Slight mistake in equation (1) (or maybe it's supposed to be a narrative decision?), the inner product of Av_i with itself should end up being s_i^2 and not 1. The vectors u_i are orthonormal, but Av_i = s_i u_i, not just u_i.

New comment by traes in "Deriving the SVD (Single Value Decomposition) from scratch"

traes — Wed, 01 Jul 2026 05:33:29 +0000

> Same transformation. Same map on every actual vector. But now it looks like an arbitrary tangle of scaling, shearing, and sign flips — you’d never guess it was just “stretch x by 3.”

I have a very hard time imagining a human writing this if they weren't intentionally attempting to sound like an LLM.

New comment by traes in "Amateur armed with ChatGPT solves an Erdős problem"

traes — Sun, 26 Apr 2026 08:08:53 +0000

Discussed at the time: https://news.ycombinator.com/item?id=47774494

New comment by traes in "Amateur armed with ChatGPT solves an Erdős problem"

traes — Sun, 26 Apr 2026 08:02:49 +0000

Exceedingly unlikely. This was one of the more discussed Erdos problems, and multiple experts have attested to the technique's novelty. If you're referring to the lack of comments on the erdosproblems website, that doesn't really mean much. From its own blog[0], the site was only started in 2023 and only really gained momentum as a place to discuss AI solving attempts, you aren't going to see serious mathematicians discussing the problems there even if there have been significant efforts to solve it.

[0]: https://www.erdosproblems.com/forum/thread/blog:1

New comment by traes in "New study compares growing corn for energy to solar production"

traes — Thu, 23 Apr 2026 03:16:42 +0000

The reason plants capture energy from this range is because that's where most of sunlight's energy is concentrated, which is going to drop this quite a bit further. Glancing at a solar radiation spectrum curve makes it look a lot closer to ~1.5x. Combine that with inefficiencies of both the panel and the LEDs and it really doesn't look that good.

New comment by traes in "Not all elementary functions can be expressed with exp-minus-log"

traes — Wed, 15 Apr 2026 07:25:07 +0000

abs(x) = sqrt(x*x), no?

New comment by traes in "All elementary functions from a single binary operator"

traes — Mon, 13 Apr 2026 05:08:30 +0000

This preprint was written by a researcher at an accredited university with a PhD in physics. I'm sure they know what a vector valued function is.

The point of this paper is not to revolutionize how a scientific calculator functions overnight, its to establish a single binary operation that can reproduce the rest of the typical continuous elementary operations via repeated application, analogous to how a NAND or NOR gate creates all of the discrete logic gates. Hence, "continuous mathematics" as opposed to discrete mathematics. It seems to me you're being overly negative without solid reasoning.

New comment by traes in "All elementary functions from a single binary operator"

traes — Mon, 13 Apr 2026 04:58:00 +0000

Elementary function is a technical term that this paper uses correctly, not a generic prescription of simplicity.

See https://en.wikipedia.org/wiki/Elementary_function.