mathnerd314

Wiki Contributions

Comments

Sorted by

So the rent it pays is something like when you put up a sign "DO NOT DISTURB - AUTOMATED GUN TURRETS INSIDE" on your apartment. It doesn't change the fact that you have to pay rent, or the dollar amount, but it means you can spend less effort on negotiating late payments, because the landlord will wait just a bit longer before kicking you out.

In terms of anticipation, it's something like "if I did it over again, I'd still have done the same thing, because my behaviors were dictated by circumstance". I guess it's sort of like timeless decision theory.vs causal decision theory; in TDT, you can trust that your past self has modified you to make the correct decision in the future, while a causal decision theorist would not trust such an assumption.

I don't think it's pretending. They are both human, after all. Explaining away the difference in statements as situational bias (math is unpopular, so Grothendeick perceived it as hard; business is popular, so Jobs perceived it as easy) does not seem irrational to me. If we dissolve the question of blame, then why should praise be any different?

I think you've pinpointed the difficulty I ran into before I decided to just post it; Grothendieck was self-effacing, while Jobs was self-aggrandizing, and there isn't really enough in common for Jobs to say something that would present him in the same light as Grothendieck. Even the quotes I did have, from other people, were kind of stretching the bounds of credulity (e.g., the Samsung quote is preceded by 'the consumer perspective is that...', and then he goes on to talk about re-marketing android as being better than apple). I guess was trying to compensate for the 'worse-than-average effect' by using a different industry, but there should be a better/quicker way to recalibrate your self-image.

Steve Jobs, and life/death in general. Nothing too serious, just a prompt for discussion.

My post was not really designed to be followed, but more to use the collective makeup of LW as a human computational cluster / search engine / associative memory. I actually got a real response (ChristianKI), which I'm very happy about. I guess SolveIt can ban me if he really wants. (A guy named 'SolveIt'. People make themselves less human supposedly to benefit others - is it artificial intelligence or just people pretending they're intelligent? is gwern a robot? o_O)

The last paragraph was just brain-dumping my expectations for the conference. I was a bit off the mark. No sex and very little music, although there was a lot of alcohol and people shouting at each other. I'm guessing no LW'er besides me would consider it anything other than a waste of time.

There was also the question before Jobs's quote: when can an AI self-modify safely, or when can a human legally do mind-altering substances? Maybe there was an answer in the AIXI series; I didn't really get an answer at the conference, although I did see various effects of alcohol first-hand.

I was there 2005-2007.

I just rewrote your post to be about me / Steve Jobs instead of you / Grothendieck, see http://lesswrong.com/r/discussion/lw/lpp/a_long_comment/. Maybe you can understand what I mean about the post "mostly not being about math".

As a former (3-time) MathPath student, I have the feeling I've seen you before. I must admit that it's only a feeling.

As far as Grothendieck goes, I think he is simply channeling Buddhism's concept of beginner's mind. Nothing new, really. Most quotes are null-content "yes I'm a human" type things. The main problem I have with your post is that none of it is math-specific; take out the "math" repetition, the few mentions of calculus etc., and it's simply a generic description of ability.

And UnBBayes does computational analyses, similar to Flying Logic, except it uses Bayesian probability.

There not good evidence for the claim that reading a list of a bunch of biases improves your decision making ability. See Eliezers discussion on the hindsight bias: http://lesswrong.com/lw/il/hindsight_bias/

I checked that the procedure accounts for the biases. Hindsight bias is avoided by computing uncertainty using a regression analysis. Availability bias is avoided by using a large database with random sampling. Etc. I haven't gone through all of them, but so far the biases I've looked at can't affect the decision outcome because the human isn't directly involved in those stages of computation.

Someone wearing a black belt is probably going to be perceived as more aggressive

And there's even a study on black uniforms that shows they increase perceived aggression.

Changes in confidence and body language.

This page says martial arts training increases dominance, as you say. On the other hand, that study also says that martial arts training decreases (observed) aggression. This study says perceived aggressiveness is highly correlated with proportion of mixed-martial-arts fights won, which I interpret as also meaning that martial arts training increases perceived aggression before a fight (since martial training ought to result in winning more martial arts fights). So it looks like martial arts training encourages controlling the aggressiveness signal, suppressing it in some non-fighting cases and enhancing it in competition. Or else the actual aggression levels decreased because the willingness to fight was communicated more clearly and thus people chose to fight less because their estimates of the costs rose.

We went through many separate points and at the moment I don't know how to pull them in a good way together into one post. If you see a decent way feel free.

My general writing strategy is as follows: I go through source material, write down all the quotes/facts that seem useful into a bullet list, then sort alphabetically, then reorder and group the bullets, then rewrite the sub-bullets into paragraphs, then reorder the paragraphs, then remove the list formatting and add paragraph formatting, then add a title and introduction. (The conclusion is just more facts/quotes). I've practiced this on a couple of my required-because-core essays and they've gotten reasonable marks (B+ / A- level depending on how nice the teacher is).

The problem is that you assume that you know the relevant biases.

Wikipedia has a list; I've checked a few of them, and the rest are on my TODO list. I have that page watched so if there's a new bias I'll know.

There are often cases where you don't know why someone screws up. There are domains where it's easier to get knowledge about how much people screw up than understanding the reasons behind screwups.

Information is produced regardless, and often recorded (see e.g. Gwern's Mistakes page). So long as I myself don't screw up, which, assuming that I always follow my decision procedure and my decision procedure is correct, I won't, then it doesn't matter.

Fear produces fight or flight responses. People often fight out of fear. Aggression often comes out of weakness.

OK, but I was talking about "perceived willingness to be aggressive" (signal), not aggression (action).

A karate black belt is dominant but usually not aggressive. Taller people get payed more money because being tall is a signal for social dominance.

Someone wearing a black belt is probably going to be perceived as more aggressive, the same way someone idly cleaning their fingernails with a sharp knife might be. Similarly if a person adopts something recognized as a fighting stance. Not certain about tall people, that's probably something else besides perceived aggressiveness, e.g. "My parents were rich and could feed me a lot".

This has gone on long enough that it might be worth summarizing into a post... do you want to write it or should I?

Load More