🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

One thing I haven't seen anywhere is a general applicative approach to machine learning. Has anyone tried doing simple MLP not with matrices, but with tree-shaped layers instead?

Russia is closer to losing than Ukraine will ever be. Just stop throwing Putin lifelines with negotiations and diplomacy he only exploits for time. Crushing sanctions on Russia, overwhelming military support for Ukraine—both should have been in place from the start.

This isn’t a budget, it’s a bunker. It’s pre-collapse preparation for violence at home and abroad, not shared prosperity. It’s a visual of MAGA paranoia, alt-right fear-mongering, and the structural abandonment of the social contract.

do you think a 4th century chinese alchemist would even be able to process what's actually necessary to achieve what he spent his entire life trying to do

Trump could also be in Kyiv, making historic decisions and selling weapons to Ukraine and the EU to help them fight against Russia, North Korea, Iran and China. But he’s not, because he listened to the wrong people. One of the biggest blunders in modern geopolitical history.

Much of mathematics is carefully constructing your own nose, so that you can follow it later

John Carmack Now that people aren't going write their code themselves, maybe it can even be Lisp. AIs don't need syntax that makes source code look accessible. They're fine programming in s-expressions.

n-category theorists can think in ways that we mortal mathematicians cannot. 😔

[Via jeanas.bsky.social on the non-Musky place.] And yes, this monstrosity is an actual commutative diagram from an actual math paper: “Comma 2-comonad I: Eilenberg-Moore 2-category of colax coalgebras” by Igor Baković arxiv.org/abs/2505.00682 (on page 53).

Someone with both a conscience and a spine, clearly unqualified for the Trump administration.

You would hope that a lawmaker would know that the Supreme Court exists to defend the Constitution and the rule of law, not the popular will. Saying the will of the people trumps the law of the land is authoritarian demagoguery.

It brings me no joy to say this but I've been vindicated once again.

Wait a minute… Trump’s mad that Walmart is blaming tariffs—that he implemented—for higher prices… and says the real issue is they don’t want to eat the cost? So the problem isn’t the tariffs. It’s that corporations won’t sacrifice their profits to cover for his economic

Functional programming is getting to be very much like AI in the sense that whenever some innovation championed by functional programmers enters the mainstream it stops being "Real FP" so people can keep saying that FP is academic and unusable

It's very useful to practice thinking about elementary set theory *without* thinking directly about elements. Spend some time thinking about sets *only* in terms of functions. Any function `f : A -> B` cuts an "A-shaped picture" out of B. In the special case where A is a

i still can't get over how good sai's monad tutorial website is and not just because i now have a banging migraine that won't go away for 6-9 business days

Local thief wants a pledge that these specific houses never have burglar alarms or dogs.

Someone wearing a bulletproof vest is only a "fair concern" to you if you’re planning to shoot them.

Calling c the "speed of light" completely misses the point. Rather, c is the "spacetime exchange rate": how many units of space you can exchange for one unit of time. In actuality, everything travels at the "speed of light", just not necessarily through space alone... (1/4)

It may be time to develop AI programming languages. Code generation must be optimized for guiding models in exploring solution space and ensuring correctness, not for human comprehension. Code specification must optimize synchronization between human intention and AI