Dacyn(David Simmons)

• Dacyn 6 Aug 2022 20:35 UTC

  1 point

  0 ∶ 0

  in reply to: Screwtape’s comment on: Cal­ibra­tion Trivia

  Sure, that patch wouldn’t have the problem I described.

  Anyway, do whatever works for you—if you find this exercise helps people train their calibration, then I suppose that’s a good thing. I guess my main point would be not to take too seriously what this method tells us about who is “best” at calibration—and I guess you’re saying people already don’t take seriously in the case of someone who is doing badly at the trivia portion, but I think the failure mode is a bit more general than that. Anyway, I guess it doesn’t matter too much.

• Dacyn 5 Aug 2022 16:55 UTC

  1 point

  0 ∶ 0

  on: Cal­ibra­tion Trivia

  The people with the best calibration scores will not be those with the most skill at calibration. It will be those who “don’t guess” on the trivia questions—they either know it or they don’t (100% of 0% chance of getting it right). This is because if you guess and have (e.g.) a 50% chance of getting it right, then even if you are perfectly calibrated about that 50%, you will still get a Brier score of 0.25, as opposed to a score of 0 for someone who “doesn’t guess”.

  Consequently, I don’t really see this game as being very useful at measuring calibration.

• Dacyn 2 Aug 2022 20:11 UTC

  3 points

  on: In­for­ma­tion is a Coun­ter­fac­tual Property

  The problem I see with this proposal is that the concept of “information” that you are using here is not a physical concept but an epistemological one. The lamp is information to your friend, but it would not be information to another person in the same location who was unaware of your plan. So trying to reduce this concept of information to fundamental physics seems incorrect—except in the sense that we can reduce your friend’s brain to fundamental physics.

• Dacyn 2 Aug 2022 19:58 UTC

  1 point

  0 ∶ 0

  on: Is there any writ­ing about prompt en­g­ineer­ing for hu­mans?

  This may be missing the point, but when talking about a population of 600 people, “This intervention will save 200 lives” and “This intervention will result in 400 deaths” actually mean different things. The former means that the number of deaths will be decreased by 200 if you do the intervention, whereas the latter means that the number of deaths will be increased by 400 if you do the intervention. Relative to not doing the intervention. I realize you meant it to be “relative to everyone dying” and “relative to no one dying”, respectively, but that is not what the sentences mean in English if you interpret them naturally.

• Dacyn 25 Jun 2022 15:18 UTC

  1 point

  in reply to: benjamincosman’s comment on: In­tel­li­gence in Com­mit­ment Races

  Sure, but Bot is only being adversarial against Alice, not Eliezer, since it makes the decision to precommit before learning whether its opponent is Alice or Eliezer. (To be clear, here I just use the shorthands Alice = less sophisticated opponent and Eliezer = more sophisticated opponent.) To put it differently, Bot would have made the same decision even if it was sure that Eliezer would punish precommitment, since according to its model at the time it makes the precommitment, its opponent is more likely to be an Alice than an Eliezer.

  So the only motivation for Eliezer to punish the precommitment would be if he is “offended at the unfairness on Alice’s behalf”, i.e. if his notion of fairness depends on Alice’s utility function as though it were Eliezer’s.

• Dacyn 11 Apr 2022 9:44 UTC

  1 point

  in reply to: logicnazi’s comment on: Mak­ing His­tory Available

  I would say that literature does in fact give insight, but that it is a different type of insight that a historical account is meant to give. For a piece of literature shows you how your emotions will be affected by hearing or seeing different circumstances; it allows your brain to develop causal models about how emotions are generated. History lessons do not do this, at least not in the same way; thus literature fulfills an important role that history is missing.

[Question] Max­i­mize in a limited do­main. Hope for the fu­ture.

• Dacyn 16 Feb 2022 18:59 UTC

  1 point

  in reply to: dynomight’s comment on: Ob­ser­va­tions about writ­ing and com­ment­ing on the internet

  Ah, I see, fair enough.

• Dacyn 15 Feb 2022 20:44 UTC

  1 point

  in reply to: Zvi’s comment on: Rock is Strong

  Probably just because its definition is simpler than logarithmic scoring. Do you think it is obvious that logarithmic scoring is better? That doesn’t seem obvious to me.

• Dacyn 15 Feb 2022 18:57 UTC

  11 points

  on: Rock is Strong

  -”If all that The Rock is cooking is setting the probability of every possible change to epsilon, then when the first of those events happens his Briar score is suddenly going to explode and he is going to lose all his Bayes points.”

  I think you are thinking of logarithmic scoring. With the Brier score, a wrong 100% prediction is scored the same as just four (wrong or right) 50% predictions, hardly an “explo[sion]”.

• Dacyn 15 Feb 2022 18:40 UTC

  4 points

  on: Ob­ser­va­tions about writ­ing and com­ment­ing on the internet

  I don’t understand your Monty Hall example. If Monty always reveals the leftmost non-car door that you didn’t pick (which I guess is what you mean), then he will reveal either (A) the leftmost door that you didn’t pick or (B) the rightmost door that you didn’t pick. In case (B) you can be sure that switching will result in a win, because if Monty passed over the leftmost door then it must be because it contained a car. On the other hand, in case (A) the chance is actually 50% that you win by switching, because Monty’s algorithm gives no information as to whether the rightmost (unrevealed) door contains a car. In both cases the conclusion is different from the standard (random) Monty Hall problem, where you have a ²⁄₃ chance to win by switching. What am I missing?

• Dacyn 12 Feb 2022 16:11 UTC

  2 points

  in reply to: Vanilla_cabs’s comment on: To Change the World

  I think Doug is making the assumption that the next in line is less likely to donate than Dr. Connor would be.

• Dacyn 6 Feb 2022 0:32 UTC

  2 points

  on: Anti-Parfit’s Hitchhiker

  Sure, but this example is so trivial that it makes me think you haven’t fully understood the point of the Parfit’s Hitchhiker scenario. The point is that a CDT agent regrets its choices even when the entire setup is known to the agent beforehand, and even though the choices are the only thing that determine the outcome, not the CDT agent’s internal decisionmaking process. A TDT agent will never regret its choices given those constraints. So if you find those constraints to be aesthetically pleasing, you will find TDT to be aesthetically pleasing.

• Dacyn 14 Jan 2022 15:47 UTC

  1 point

  on: Do­ing “good”

  And so, knowing that the true cumulative impact of a donation is never certain, why did I still choose to donate?

  Why did you phrase it like this—what does being certain about the cumulative impact have to do with the choice of whether to donate or not? Why isn’t the expected impact more important?

  This is a serious question, I have the same psychological issue where certainty of a positive outcome is sometimes more important than positive expected value. But I don’t yet have a rule which allows me to predict when this psychological effect will come into play.

• Dacyn 13 Jan 2022 18:10 UTC

  1 point

  in reply to: Midnight_Analyst’s comment on: On Virtue

  I want to distinguish this mental action from the behaviours that result from it. I’m trying not to make claims directly about who and what should be outwardly praised.

  I totally agree that this is an important distinction (it is the distinction the post I linked to is about), and when I talked about Virtue Points in my comment I was meaning the mental action. But mental actions still have effects, which is why the first sentence of my comment isn’t self-contradictory.

  On the connection to involuntary suffering, fair enough, but then you shouldn’t call it “Virtue Points”. That already means something else.

• Dacyn 13 Jan 2022 13:35 UTC

  1 point

  on: On Virtue

  From the perspective of the collective, the point of awarding Virtue Points is so that people know what traits to signal to remain in good graces with the community. From the perspective of the individual, a lot of the time that will feel like doing the Right Thing and not getting rewarded, due to phenomena discussed here.

  Since involuntary suffering doesn’t show up in this paradigm, I think it is irrelevant to the notion of Virtue Points.

• Dacyn 7 Jan 2022 17:03 UTC

  1 point

  in reply to: Valentine’s comment on: Sig­nal­ing isn’t about sig­nal­ing, it’s about Goodhart

  Hmm, I am trying to see if it is really the same puzzle?? The self-deception I see, since if you get the opposite of whatever you choose then it motivates you to self-deceive so that you’ll choose the opposite of whatever you want to get. But then why is the alternative death? Ah well, maybe it’ll make sense to me later.

• Dacyn 7 Jan 2022 14:48 UTC

  4 points

  on: Rad­i­cal open­ness—say things that oth­ers strongly dislike

  I think having the confidence to express what you are actually thinking gives a big advantage. Such a person is able to see more. They can think about uncomfortable and inconvenient ideas that most people would instinctively self-censure.

  The exact opposite is true. If you commit to yourself to not immediately taking any thought you have and shouting it from the rooftops, this gives you the mental space to think “uncomfortable and inconvenient ideas that most people would instinctively self-censure”. It’s just that once you have these ideas in your thought-space, you have to think more about how to express them than just “put them in my latest song”. Good secrets take a while to express properly.

• Dacyn 7 Jan 2022 14:22 UTC

  1 point

  on: Sig­nal­ing isn’t about sig­nal­ing, it’s about Goodhart

  This is all true.

  And yet.

  Very rarely, but I would guess at least once in your life, you will be faced with a decision whose outcome is so important that all of this is stripped away, like rationalization. At which point you are faced with the decision: to signal, or not to signal. But it is not clear which choice corresponds to which outcome: does treating it as a signal correspond to signalling more, or to signalling less, as suggested by Dagon’s comment?

  Would you rather be trustworthy, or trusted?

  The OP suggests that maybe we can have both, but what if that’s not always the case? And what if the outcome you get, is the exact opposite of the outcome you choose?

  I have no good solutions here, my stopgaps are “double-check important decisions through people you trust” and “cooperate with yourself, even if you seem like a selfish person”.

• Dacyn 1 Jan 2022 11:26 UTC

  1 point

  in reply to: Said Achmiz’s comment on: Firm­ing Up Not-Ly­ing Around Its Edge-Cases Is Less Broadly Use­ful Than One Might Ini­tially Think

  OK, fair enough. So you are asking something like “is it ever ethical to keep a secret?” I would argue yes, because different people are entitled to different parts of your psyche. E.g. what I am willing to share on the internet is different from what I am willing to share in real life. Or am I missing something again?