michaelsullivan

Karma: 237

michaelsullivan 31 Mar 2009 15:32 UTC
31 points
in reply to: bbarth’s comment on: Money: The Unit of Caring
Julian’s comment is on point though. I’ve been involved with any number of charitable organizations where it is expected that people donate significant time for things like bake sales or craft fairs or dinners in order to raise money, where if you took the money raised minus costs divided by the total hours spent, people would have done better taking second jobs at McDonald’s and donating the money.

Plus, we’re often providing a product which wouldn’t sell for that price on the open market, with custom driven largely by people’s affinity for the organization raising the money.

All in all, fund-raisers that aren’t either a good leisure activity for all involved, or relentlessly and professionally focused and profitable (i.e. don’t encourage random volunteers—only those with relevant marketable skills and make sure the venture would at least be break-even if you accounted for fair value of labor) are just a horrendous waste of resources. Just get people to write checks.

And yes I beat this drum at every socially appropriate opportunity for every charitable organization I’m associated with.

michaelsullivan 22 Apr 2009 21:46 UTC
3 points
in reply to: Annoyance’s comment on: Well-Kept Gardens Die By Pacifism
I would expect the potential commentariat at Less Wrong to be terribly small if anyone holding a firm belief that is not rationally justifiable were banned.

I am highly skeptical that I have fully purged myself of all beliefs where I have been presented with correct damning evidence against them. If anything, reading here has raised my estimate of how many such beliefs I might hold. Even as I purge many false propositions, I become aware of more biases to which I am demonstrably subject. Can anyone here who is aware of the limitations of our mental hardware say otherwise?

I am not as convinced as most posters here that all possible versions of theism are utterly wrong and deserve to be accorded effectively zero probability, but in any case, it’s clear that LW (and OB) communities generally wish to consider the case against theism closed. To the extent that the posters do not attack theism or theists in an obviously biased way, I have respected the decision and post and vote accordingly, including downvoting people who try to reopen the argument in inappropriate places.

I also intend to make a habit of downvoting those who waste time denouncing theism in inappropriate contexts or for specious reasons having more to do with signaling tribal bonds than with bringing up any new substantive argument against theism.

I don’t recall who suggested that we need another canonical example of irrationality, but I agree wholeheartedly. In fact, I’d suggest we need a decent short list to rotate through, so that no one topic gets beaten up so consistently as to encourage an affective death spiral around it.

michaelsullivan 22 Apr 2009 21:59 UTC
3 points
in reply to: Annoyance’s comment on: Well-Kept Gardens Die By Pacifism
My downvote (along with most others I presume) is not about agreement, but about whether you are adding anything useful to the discussion. Argument by repeated assertion is not supposed to be a staple of rationalist discourse. Either it’s worth your time to provide some links to an actual argument or it isn’t.

Do you really expect points for needing to get in the last word?

michaelsullivan 19 May 2009 22:36 UTC
2 points
on: Bad reasons for a rationalist to lose
On your reaction to “a way to reject the placebo effect”, it’s important to distinguish what we are trying to do. If all I care about is fixing a given problem for myself, I don’t care whether I solve it by placebo effect or by a repeatable hack.

If I care about figuring out how my brain works, then I will need a way to reject or identify the placebo effect.

michaelsullivan 11 Dec 2009 21:17 UTC
1 point
in reply to: mattnewport’s comment on: You Be the Jury: Survey on a Current Event
The spread between the base prior for male on female sexual violence vs. female on female sexual violence is nowhere near enough to separate the percentages for Knox and Sollecito this widely, let alone as widely as in your first cut.

The general prior may be 2-3 times as likely, but that spread should not hold up after a lot of additional evidence to consider where they are essentially equal. Every piece of evidence that counts for or against them roughly equally should shrink that spread in ratio terms. And it should not expand it in absolute terms at all either. Given that these priors are extremely low (or should be), well under 1% for male or female, there shouldn’t be more than a 1% spread in those simply from that prior. Only evidence which would amplify the male v. female disparity in prior, or evidence that specifically argues in favor of S’s but not K’s guilt, should get you to a spread of more than 1% between them.

michaelsullivan 16 Dec 2009 14:59 UTC
−4 points
in reply to: komponisto’s comment on: The Amanda Knox Test: How an Hour on the Internet Beats a Year in the Courtroom
This is entirely wrong. The evidence against G should modify both P(Knox) and P(random peer that is not G) downward.

Proximity should keep P(Knox) > P(random peer) unless there is evidence specific to Knox which lowers her P (i.e. a good alibi).

Your statement about proceeding from the physical evidence and ignoring other things is a heuristic. The fact is that some random suspicious behavior by K is evidence against K, it is just extremely weak evidence. Let R = random suspicious behavior by K. I contend that P(K|R) > P(K|^R). Your arguments that many people do R all the time and are not murderers address the strength of the evidence, but do not address the sign, unless you wish to contend that people who engage in R are less likely or equally likely to be murderers than people who do not. You have made no such argument.

It’s clear that the evidence against K should be overwhelmed by the evidence against G, given that no solid connection was established between G and K. But it isn’t zero evidence, it is merely very weak evidence.

You have established a safety heuristic to keep yourself from overvaluing weak evidence, but your safety heuristic has it’s own shortcomings, because it has caused you to give it zero weight, which is obviously wrong.

michaelsullivan 14 Jan 2010 14:50 UTC
4 points
in reply to: Kutta’s comment on: Reference class of the unclassreferenceable
The problem with taw’s argument is not that outside view has failed, he has simply made a bad choice of reference class. As noticed by a few commenters, the method chosen to find a reference class here, if it worked, would provide a fully general counterargument against the feasibility of any new technology. For any new technology, the class of previous attempts to do what it does is either empty, or a list of attempts with a 0% success rate. Yet somehow, despite this, new tech happens.

To use outside view in predicting new technology, we have to find a way to choose a reference class such that the track record of the reference class will distinguish between feasible attempts and utterly foolish ones.

michaelsullivan 25 Mar 2010 18:06 UTC
0 points
in reply to: komponisto’s comment on: The Amanda Knox Test: How an Hour on the Internet Beats a Year in the Courtroom
since you appear to still be watching this thread, it seems worth a reply even 3 months later.

Clearly “entirely wrong” is too strong and unnecessary verbiage in any case. I admit it did not even occur to me that you were using “=” to mean only “essentially equal in absolute terms”.

Note, my comment was referring only to your response to Psychohistorian, not your general argument about why P(Knox) is very small given the information presented at trial, with which I agree wholeheartedly.

I apologize for my confusion. Here and in a few other comments, you seemed to be asserting not merely that the strong evidence against G put so much probability mass into P(G) that the differences between P(Knox) and P(random peer) became much too small to care about in absolute terms, but that it somehow changed the relative dynamic between P(K) and P(rp). The fact that neither probability was worth worrying about in legal terms doesn’t change the ‘>’ vs. ‘=’ question, and that’s the only thing I was responding to there. I also suspect the distinction matters in practice in rare cases, so I saw it as important enough to be worth picking the nit.

michaelsullivan 12 Apr 2010 17:29 UTC
0 points
in reply to: pjeby’s comment on: Pain and gain motivation

[SASS threats] Most of these threat predictions are completely bogus in the modern world.

This rings true in general, and yet I can see at least one or two negative hangups in my own life and the life of friends and family where there is a significant non-bogus threat associated with failure (or success) of a particular task.

Do you have techniques which address such situations? It would seem getting rid of the negative motivator completely could be impossible in such situations, short of delusion or intentional forgetfulness/detachment.

michaelsullivan 10 May 2010 21:45 UTC
1 point
on: Beauty quips, “I’d shut up and multiply!”
Your update doesn’t solve the problem. It’s a semantic issue about what credence we are being asked. If we are being asked about the probability of our coin flip associated with this iteration of the experiment, then the answer is ¹⁄₂. If we are being asked about the probability of the coin flip associated with this particular awakening, then it must be ¹⁄₃.

You say that you must use cell counts of 500,250,250, but the fact is that if you repeat the experiment 1000 times, sleeping beauty will be awoken 1500 times, not 1000. So what are you doing with the other 500 awakenings? I would say you are implicitly ignoring them, as you do when you say “we only accept her last decision” in the bet scenario. The reformulations of this using different people, rather than the same person being awoken multiple times don’t seem to cause as much trouble.

The semantic issue here is reminiscent of arguments I’ve seen over the Monty Hall problem when it is misstated so that Monty’s algorithm is not clear. People who assume what he usually does on the show come up with ²⁄₃, and people who don’t make any assumptions come up with ¹⁄₂ (as do most of the people who simply don’t understand restricted choice).

michaelsullivan 13 May 2010 18:03 UTC
7 points
on: Aspergers Poll Results: LW is nerdier than the Math Olympiad?
One thing I noticed on the AQ test is a number of questions which don’t clearly relate to any of the direct symptoms, but are chosen apparently because these things correlate with Asperger’s/Autism.

For instance, the statement “I find numbers fascinating”. This may well correlate with Autism, but I bet it correlates even more strongly with an interest in math or math-heavy sciences. That math olympians nearly all answered yes to this question, doesn’t really tell us anything new about the correlation between math and AQ. The fact that this question is scored positively for AQ has built the correlation assumption into the test.

There are some less obvious examples such as “I see patterns all the time”. I would wild-ass guess that at least 2-3 points of the gap between the math heavy groups’ scores and the population average scores are represented by answers to questions that are not related directly to symptoms, but are personality/brain traits correlated with symptoms, which also happen to be highly useful for doing math.

michaelsullivan 28 Aug 2010 17:43 UTC
1 point
in reply to: paulfchristiano’s comment on: Rationality Lessons in the Game of Go
Losing fights sooner is an important go lesson, related to the OPs point about invasions.

The sooner you realize that you’ve lost a fight, the less you typically lose (early detection means lots of aji).

My favorite breakthrough when I was learning go, was the game that I crushed an opponent a stone stronger than me (or rather a stone stronger than I was before the breakthrough) by giving way at every fight, and then at the end of the middle game, destroying a formation he thought was safe with the aji created along the way. It was all about paying attention to what I could gain while giving up original aims, and how that would affect situations elsewhere on the board. For a while that became my style: winning by losing: lose every fight but win the game. That was my jump from 6k to 3k overnight.

michaelsullivan 1 Apr 2011 21:39 UTC
2 points
in reply to: RHollerith’s comment on: Reflections on rationality a year out
In the present environment, at least the in the US and most of Europe, it is conceivable that counterfeiting money has positive externalities. There is a very high unemployment rate, and low capacity utilization across most sectors of the economy. There is a fairly broad school of economists who believe that this is the result of a shortage of aggregate demand brought on by poor macroeconomic management due to an irrational fear of inflation—that the central bank can and should do more than it is doing to stimulate the economy, and failing that, central goverments not facing high or rising borrowing costs should be willing to run large short-term deficits. If this bunch of economists is correct, then these policies would be good for the global economy. Since counterfeiting money is essentially equivalent to monetary stimulus, it also would have positive externality. It would be much more likely to put some resources back to work and have little or no effect on the value of assets denominated in that currency.

If all economic actors are perfectly rational, and none suffer from money illusion, hyperbolic discounting, or other effects, then you would be right in all times, not just in normal times of close to optimal fed policy and near full labor and capital usage. That would also mean that the economists to which I refer would be wrong about the current state of events.

I agree, though, that buying junk you do not want would destroy most of any utility gained by counterfeiting. It would be far better to buy things you do want, or failing that, to simply give the money away.

michaelsullivan 13 Apr 2011 23:19 UTC
2 points
in reply to: Spencer_Sleep’s comment on: Verifying Rationality via RationalPoker.com
“If you improve your rationality and knowledge of basic probability to the point where it exceeds that of the average at the table you are playing at, you will (on average) make money.”

Only if you are playing in an unraked home game.

In venues where you play for significant amounts of money versus strangers, there will be a house guaranteeing the fairness of the game and providing insurance against stealing, etc. and they collect a lot of money for this service relative to what a good player can expect to win. Unless you play nosebleed stakes (where the house can make a lot of money by taking a very small percentage of the pot), the rake will make somewhat above average players losers, and below average players big losers.

In a typical low-limit game, the very best players will net on average about as much as the house is taking from each player. So you have to be about halfway from average to the best to break even.

michaelsullivan 19 Apr 2011 17:37 UTC
10 points
0
in reply to: SilasBarta’s comment on: Learned Blankness
It looks like that formula is a lot like cutting the ends off the roast.

The answer to “who cares?” is most likely “some 1930s era engineer/scientist who has a great set of log tables available but no computer or calculator”.

I am just young enough that by the time I understood what logarithms were, one could buy a basic scientific calculator for what a middle class family would trivially spend on their geeky kid. I remember finding an old engineer’s handbook of my dad’s with tables and tables of logarithms and various probabilistic distribution numbers, it was like a great musty treasure trove of magical numbers to figure out what they meant.

I don’t know where that ended up, but I still have his slide rule.

Of course, even in the day, it would make more sense to share both formula, or simply teach all students enough math to do what Gray does above and figure out for yourself how to calculate the model-enlightening formula with log tables. Since you’d need that skill to do a million other things in that environment.

michaelsullivan 24 Apr 2011 1:07 UTC
19 points
in reply to: brazil84’s comment on: Epistle to the New York Less Wrongians
The disadvantage of excluding women (or men) is far too large. Just like any other rationalist, they have information, experience and perspective that is valuable. And more so than rationalists of the same gender as you, they can share near insight about issues personal to women and far insight into issues personal to men, that is extremely rare to find among men or a group of men. There is a whole realm of gender related affective death spirals that are terribly easy to fall into in segregated gender groups. This applies to almost any other significant culture gap as well (black/white, rich/poor urban/rural etc.).

There’s a common error described some places as “privilege blindness” referring to how easy it is for those who are privileged in some way to go through life completely oblivious to the way the world works for those who do not share that good fortune. This is a classic example of an affective death spiral, and will be a huge potential pitfall for any all-male or even mostly male group.

It might make sense for larger groups with plenty of both genders to have some separate meetings, and that’s a worthy experiment. But keeping apart indefinitely seems extremely unwise.

Parent upvoted even though I disagree strongly, because this is an issue worth discussing and bringing in empirical data.

michaelsullivan 25 Apr 2011 21:10 UTC
0 points
on: New Haven / Southern Connecticut Meetup, Wednesday Apr. 27th 6 PM
I’d be interested but can’t make it. I’m rehearsing that night for a concert on the 30th with New Haven Oratorio.

michaelsullivan 25 Apr 2011 21:37 UTC
3 points
in reply to: SilasBarta’s comment on: Mini-camp on Rationality, Awesomeness, and Existential Risk (May 28 through June 4, 2011)
whatever length they choose, it will be too long or too short for many/most things/people.

1 week is already a long time to devote if you are already focused on a career. I couldn’t realistically do even the mini-camp without a lot of advance planning. OTOH, If you are still in school or have an academic job that gives you summers off, then 10 weeks is equivalent to a summer internship or fellowship. I’m not sure what the medium would be except maybe 2 weeks—harder but still doable for those who don’t have summers off. Once you go longer than 2 weeks, it’s almost impossible for anyone with serious family or career commitments other than education/academe, so you might as well do the whole summer if it makes sense.

michaelsullivan 26 Apr 2011 17:48 UTC
2 points
in reply to: Eliezer Yudkowsky’s comment on: Zut Allais!
“What the coherence proofs for expected utility show, and the point of the Allais paradox, is that the invariant measure of distance between probabilities for this purpose is the usual measure between 0 and 1. That is, the distance between ~0 and 0.01, or 0.33 and 0.34, or 0.99 and ~1, are all the same distance.”

In this example. If it had been the difference between .99 and 1, rather than ³³⁄₃₄ and 1, then under normal utility of money functions, it would be reasonable to prefer A in the one case and B in the other. But that difference can’t be duplicated by the money pump you choose. The ratios of probability are what matter for this. ³³⁄₃₄ to 1 is the same ratio as .33 to .34.

So it turns out that log odds is the right answer here also. If the difference in the log odds is the same, then the bet is essentially the same.

michaelsullivan 27 Apr 2011 17:51 UTC
0 points
in reply to: michaelsullivan’s comment on: New Haven / Southern Connecticut Meetup, Wednesday Apr. 27th 6 PM
In fact, I probably can swing by for a short time, will just have to take off at 6:45, and may not get there by 6, but I’ll give it a shot.