on reddit there was a funny, you would write !remindMe x days/years and then reminder bot would come after this time and reply
Warty
Karma: 198
Math contains a lot of stuff
Zack Davis (the wise): 20 years ago I read in the Sequences that P[link]
For example, if I said to my wife “I think there’s an 8% chance we’ll get divorced in the next five years,” the very act of sharing that prediction might cause that number to go up! She might reasonably interpret it as a sign that I don’t believe in the partnership. Then, she might respond with her own prediction of 14%, leading me to update in the same way, in a back-and-forth pattern that ends up in divorce. This cascade of updates is only irrational in that speaking purely in terms of high-level probabilities is the wrong tool for the job of coordination
I don’t have the math proof at hand but, I would figure that it is also irrational in another way. If both people are perfectly rational and honest, I expect that communicating wouldn’t make them both worse off.
Does structuralism hold that the statement of the Continuum Hypothesis has a truth value? If no, how does it differentiate between my hash of BB mod 2 statement and CH?
It suffices for me if there is one philosophy of mathematics that exhibits Less Wrongist virtues 🙂
The Yud view seems not very complete from this. I’m saddened to see I can’t access the Q&A videos, but I found a transcript of #14 about infinite atheism here. Note that the most infinite thing in my example suspicious statement is quantification over natural numbers.
I like both your posts, thanks for summarizing these views. I think in combination it somewhat dissolved the philosophical mystery for me, but there is still some normal empirical mystery left. The Mac Lane stuff reminded me of something I read at Terence Tao’s, about how mathematicians use different intuitions like geometric, economic.
[Question] What’s the LessWrongist philosophy of mathematics?
Isn’t “this program is thinking about X” a kind of semantic property?
By the definition used in the theorem, no. Semantic means input/output.
Rice’s theorem has low applicability to real life. You rarely need to take a completely arbitrary program as input, and when you do, non-semantic properties (e.g. this program does something within the first 10^9999 years of running it) are more interesting to you than semantic ones.
“Turing-completeness” is something of a pop-sci notion with no one precise meaning. Neural networks as used in LLMs can’t loop forever, which makes Rice’s theorem not apply to them.
Maybe the real anti-violence is to retrospectively support any violence from your side, because that would hurt the cause and thereby de-incentivize violence 🤔
Statistics show that civil movements with nonviolent doctrines are more successful at attaining their stated goals (especially in states that otherwise have functioning police). The factions that throw away all their morals lose the sympathy of the public and politicians, and then they fail. Terrorism is not an instant ‘I win’ button that people only refrain from pressing because they’re so moral. Society has succeeded in making it usually not pay off—say the numbers.
What are the statistics? I’m not convinced. It seems the sympathy part is mostly solved by having a large official wing that disavows any violence.
Directly stopping all AI is not the only way terrorism could raise P(pause). It also raises awareness of the cause, and causes terror which can change behavior.
Yes, it’s mysterious. With a singleton AI, selection doesn’t apply. So we could only say like, in the long run the replication-only aliens occupy more space than the side project ones. (Naively, it would seem you could still at least “defend your territory” if you spend 1% on fun.)
Still, when I hear speculations of what unaligned ASI will be doing it doesn’t mesh. Replicator feels to me an obvious default, some simplicity intuition as they call it.
Like if we imagine the universe across time, feels clean if the right part of the graph is “dumb superintelligent replicator era”
standard yudkowskian belief is that superintelligence will “fill the lightcone” with something (paperclips if bad, people if good). seems sus to me, I would expect the law to be that life just fills space with itself asap. e.g. animals don’t make paperclips, just make themselves
It feels deeply uncomfortable to be participating in an elite AI x-risk fellowship and tell your peer, manager, or mentor: “idk why ASI poses an existential risk.”
With some nuances, if you don’t know why ASI poses x-risk, a better word choice might be “idk if ASI poses x-risk”.
That frame of mind might be good for successfully executing strategies 1 and 3.
Persuasion. Getting people to care about it.
I thought persuasion was when you convince someone of something. It’s often an additional step after informing them of what you think is true.
In my experience, Persuasion is literally the easiest step.
I’m not convinced. Yud’s book came out months ago and it seems many important people have read it. I would think if the persuasion rate (of the top people) was not very low, things would look much different now.
Let me raise an inconvenient world for you: there are two types of actions in politics, fake and real. For fake actions such as signing a vague statement, attention/information might suffice. For real actions such as passing a law that will prevent lots of money from being made, persuasion of the underlying worldview is necessary.
I happen to think we do live in this world 🙂
People don’t express anger and hatred for no reason. When they do, it’s because they have reasons to think something is so bad that it deserves their anger and hatred.
That’s often false, people also express anger and hatred to:
get attention
make the target look bad and themselves look good
activate lizard brain in the audience
There are in fact norms agains some (genuine or not) expressions of anger. The question of applying norms is another nontrivial question. Which is why Goofusia’s take is plausible (what she is exploiting if you will).
Yudkowskian self help thought is weird such as the “firewall” (between beliefs and what would be good to believe)
But I think my mind doesn’t work with the firewall, for example if I slightly delulu myself that my efforts will work out I have energy and such, and idk how to get these benefits with the firewall.
I recall it was a big problem for me for a few weeks when I was reading hpmor
propaganda (in its literal meaning, “public communication aimed at influencing an audience and furthering an agenda”)
Wiktionary at the link marks that usage as dated, I wouldn’t call it “its literal meaning”. I believe the closest current single word would be “advocacy”.
I see how it makes sense but it was super freaking hard for me to understand why “Cassandra/Mule”. These words mean so many things I guess
Real-world disagreements tend to persist; they’re predictable—in flagrant contradiction of how the beliefs of Bayesian reasoners would follow a random walk. From this we can infer that typical human disagreements aren’t “honest”, in the sense that at least one of the participants is behaving as if they have some other goal than getting to the truth.
Technically, it’s sufficient if one of the participants incorrectly believes this about the other. If I incorrectly believe you’re a bus arrival time scammer, we won’t converge despite both being honest.
your MO link appears to be broken
I think I covered that in my math A definition with “motivation from [...] other math”, no?
My examples show that math C in the limit separates from A-B. Factoring could be part of some math A-B but is not in itself. Practical algorithms are somewhat similar to theoretical algorithms, but it’s not a question that would be asked in math A, they only care about behavior in the limit (leading to galactic algorithms). And the AI development example is different from anything in math A.
It’s a good point that A and B are intertwined but I don’t think it literally all goes back to Euclid. (even then, at least Euclid would be the one mathemAtician 🙂). I think information theory and game theory are examples coming out of application and not other math? CS sort of came out of logic, but before figuring out computation the problems weren’t mathematically specified.