Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Comment author: dxu 04 April 2015 02:21:29AM 0 points [-]

Another way of viewing this would be that my preferences run thus: (D,C);(C,C);(C,D);(D,D) and Clippy run like this: (C,D);(C,C);(D,C);(D,D).

Wait, what? You prefer (C,D) to (D,D)? As in, you prefer the outcome in which you cooperate and Clippy defects to the one in which you both defect? That doesn't sound right.

Comment author: lerjj 06 April 2015 10:25:16PM 0 points [-]

woops, yes that was rather stupid of me. Should be fixed now, my most preferred is me backstabbing Clippy, my least preferred is him backstabbing me. In the middle I prefer cooperation to defection. That doesn't change my point that since we both have that preference list (with the asymmetrical ones reversed) then it's impossible to get either asymmetrical option and hence (C,C) and (D,D) are the only options remaining. Hence you should co-operate if you are faced with a truly rational opponent.

I'm not sure whether this holds if your opponent is very rational, but not completely. Or if that notion actually makes sense.

Comment author: VAuroch 18 November 2013 10:44:44AM 4 points [-]

It is a personal peeve when any explanation of the Bell Inequality fails to mention the determinist Big Loophole: It rules out nearly all local hidden-variable theories, except those for which the entire universe is ruled by hidden variables. If you reject the assumption of counterfactual definiteness (the idea that there is a meaningful answer to the question "what answer would I have gotten, had I conducted a different experiment?"), local hidden variable theories are not ruled out. This leads to superdeterminism and theories which assume that, through either hidden variables stretching back to t=0 or backwards-in-time signals, the universe accounted for the measurement and the result was determined to match.

This is, in fact, what I held to be the most likely total explanation for years, until I better understood both its implications and MWI. Which, in fact, also rejects counterfactual definiteness. MWI does it one better; it rejects factual definiteness, the idea that there is a well-defined answer to the question "What answer did I get?", since in alternate worlds you got different answers.

Comment author: lerjj 06 April 2015 10:08:28PM 1 point [-]

Sorry for being a pain, but I didn't understand exactly what you said. If you're still an active user, could you clear up a few things for me? Firstly, could you elaborate on counterfactual definiteness? Another user said contrafactual, is this the same, and what do other interpretations say on this issue?

Secondly, I'm not sure what you meant by the whole universe being ruled by hidden variables, I'm currently interpreting that as the universe coming pre-loaded with random numbers to use and therefore being fully determined by that list along with the current probabilistic laws. Is that what you meant? If not, could you expand a little on that for me, it would help my understanding. Again, this is quite a long time post-event so if anyone reading this could respond that would be helpful.

Comment author: query 04 April 2015 03:57:21AM 1 point [-]

I agree it is better if both agents cooperate rather than both defect, and that it is rational to choose (C,C) over (D,D) if you can (as in the TDT example of an agent playing against itself). However, depending on how Clippy is built, you may not have that choice; the counter-factual may be (D,D) or (C,D) [win for Clippy].

I think "Clippy is a rational agent" is the phrase where the details lie. What type of rational agent, and what do you two know about each other? If you ever meet a powerful paperclip maximizer, say "he's a rational agent like me", and press C, how surprised would you be if it presses D?

Comment author: lerjj 06 April 2015 08:37:12PM 0 points [-]

In reality, not very surprised. I'd probably be annoyed/infuriated depending on whether the actual stakes are measured in billions of human lives.

Nevertheless, that merely represents the fact that I am not 100% certain about my reasoning. I do still maintain that rationality in this context definitely implies trying to maximise utility (even if you don't literally define rationality this way, any version of rationality that doesn't try to maximise when actually given a payoff matrix is not worthy of the term) and so we should expect that Clippy faces a similar decision to us, but simply favours the paperclips over human lives. If we translate from lives and clips to actual utility, we get the normal prisoner's dilemma matrix - we don't need to make any assumptions about Clippy.

In short, I feel that the requirement that both agents are rational is sufficient to rule out the asymmetrical options as possible, and clearly sufficient to show (C,C) > (D,D). I get the feeling this is where we're disagreeing and that you think we need to make additional assumptions about Clippy to assure the former.

Comment author: query 03 April 2015 09:11:15PM 1 point [-]

I think what might be confusing is that your decision depends on what you know about the paperclip maximizer. When I imagine myself in this situation, I imagine wanting to say that I know "nothing". The trick is, if you want to go a step more formal than going with your gut, you have to say what your model of knowing "nothing" is here.

If you know (with high enough probability), for instance, that there is no constraint either causal or logical between your decision and Clippy's, and that you will not play an iterated game, and that there are no secondary effects, then I think D is indeed the correct choice.

If you know that you and Clippy are both well-modeled by instances of "rational agents of type X" who have a logical constraint between your decisions so that you will both decide the same thing (with high enough probability), then C is the correct choice. You might have strong reasons to think that almost all agents capable of paperclip maximizing at the level of Clippy fall into this group, so that you choose C.

(And more options than those two.)

The way I'd model knowing nothing in the scenario in my head would be something like the first option, so I'd choose D, but maybe there's other information you can get that suggests that Clippy will mirror you, so that you should choose C.

It does seem like implied folk-lore that "rational agents cooperate", and it certainly seems true for humans in most circumstances, or formally in some circumstances where you have knowledge about the other agent. But I don't think it should be true in principal that "optimization processes of high power will, with high probability, mirror decisions in the one-shot prisoner's dilemma"; I imagine you'd have to put a lot more conditions on it. I'd be very interested to know otherwise.

Comment author: lerjj 03 April 2015 10:11:53PM *  1 point [-]

I understood that Clippy is a rational agent, just one with a different utility function. The payoff matrix as described is the classic Prisoner's dilemma where one billion lives is one human utilon and one paperclip on Clippy utilon; since we're both trying to maximise utilons, and we're supposedly both good at this we should settle for (C,C) over (D,D).

Another way of viewing this would be that my preferences run thus: (D,C);(C,C);(D,D);(C,D) and Clippy run like this: (C,D);(C,C);(D,D);(D,C). This should make it clear that no matter what assumptions we make about Clippy, it is universally better to co-operate than defect. The two asymmetrical outputs can be eliminated on the grounds of being impossible if we're both rational, and then defecting no longer makes any sense.

Comment author: Allan_Crossman 04 September 2008 12:29:50AM 12 points [-]

Prase, Chris, I don't understand. Eliezer's example is set up in such a way that, regardless of what the paperclip maximizer does, defecting gains one billion lives and loses two paperclips.

Basically, we're being asked to choose between a billion lives and two paperclips (paperclips in another universe, no less, so we can't even put them to good use).

The only argument for cooperating would be if we had reason to believe that the paperclip maximizer will somehow do whatever we do. But I can't imagine how that could be true. Being a paperclip maximizer, it's bound to defect, unless it had reason to believe that we would somehow do whatever it does. I can't imagine how that could be true either.

Or am I missing something?

Comment author: lerjj 03 April 2015 08:23:37PM 1 point [-]

7 years late, but you're missing the fact that (C,C) is universally better than (D,D). Thus whatever logic is being used must have a flaw somewhere because it works out worse for everyone - a reasoning process that successfully gets both parties to cooperate is a WIN. (However, in this setup it is the case that actually winning would be either (C,D) or (C,D), both of which are presumably impossible if we're equally rational).

Comment author: Manfred 14 March 2015 05:22:45AM *  3 points [-]

It's from the most common spoken order. "March fourteenth, twenty-fifteen."

Comment author: lerjj 14 March 2015 11:34:07AM 5 points [-]

Are you sure that's right chronologically? Just because in the UK we use dd/mm/yy and we say "Fourteenth of March, twenty-fifteen".

Japan apparently uses yy/mm/dd which makes even more sense, but I have no idea how they pronounce their dates. Point being, I'm not sure which order things actually evolved in.

Comment author: garabik 04 March 2015 02:28:02PM 4 points [-]

I am starting to think that Lord Voldemort planned to commit suicide by proxy. Being without any personal aims, totally bored, without any happiness, surrounded by idiots, no chance of improvement - yet he cares about the world, at least somewhat, and realized that the original plan of playing chess with Harry would not alleviate his mood anymore anyway. And he has a better, happier and (age adjusted) more intelligent clone running around, so it is not like he will cease to exist altogether.

So he told Harry where to find Memory charms, prepared the plot, got the Stone (for Harry), made Harry take the Vow to keep his recklessness in limits, let him keep his wand and put him to the Final Exam.

The problem is that if he planned to be Obliviated, the plot was extremely complicated and relied on too many factors that could have gone wrong. So perhaps LV just threw the towel and said to himself, darn, let's Harry try whatever he can think about to do with us, and if he fails, well, it's not like I cannot try something else in some decades.

Comment author: lerjj 04 March 2015 10:11:14PM 1 point [-]

This would to some extent letting Harry keep his wand- he wants to have some fun after all, and Harry should be given a very limited chance to win. Not much, maybe strip him naked, surround him by armed killers and point a gun at his head, whilst giving him only a minute to think. But leave him his wand, and do give him the full 60 seconds, don't just kill him if he looks like he's stalling.

Comment author: Mperv 04 March 2015 08:41:49PM *  3 points [-]

"Other living weapons cannot be Transfigured; they will not survive the disenchantment for the requisite six hours to avoid being traced by Time-Turner."

So, how does Harry plan to evade been traced by Time-Turner?

Is it correct that Voldemort will die when he will be disenchanted?

By the way, why did Voldemort make a description of Hermione resurrection ritual and put it in a pouch if he planed to kill Harry anyway?

Comment author: lerjj 04 March 2015 10:01:15PM 3 points [-]

Well, seeing as he was almost prophesied to fail, it was sensible to make sure Harry would have someone to stop him in the future. And as it turns out, this was a very good idea.

Comment author: Sheaman3773 04 March 2015 08:35:09PM 13 points [-]

Well, that's certainly one way to explain away all of the strange aspects. Establish them as fact, through the mysterious bond between LV and HP, and do so in front of a huge crowd so that the word can spread and mutate on its own. By the time anyone comes to investigate or question, they will already be influenced by the show or rumors they've heard, promoting that hypothesis to their attention rather than coming to it naturally.

It's pointing the police at Mortimer Snodgrass, from chapter 17, as it were.

Comment author: lerjj 04 March 2015 09:57:17PM 2 points [-]

It's actually the same tactic as the Weasley twins used to cover the "engaged to Ginever Weasley" story- plant so many make newspaper reports that everyone gets confused. And it kinda happens again after the Hermione/Draco incident. Guess Eliezer like the theme of people not being able to discern the truth from wild rumours if the truth's weird enough.

Comment author: lerjj 03 March 2015 09:06:28PM 1 point [-]

So... what we should do now is to work out all the things Quirrell should have before this. He couldn't predict partial transfiguration, true. But he knew that Harry had a power he knew not, and had a long time to plan for contingencies.

Personally, I think he should have had the death eaters disillusioned, surround Harry but from a distance, cast holograms to confuse him and then use ventriliquo charms. At the very least disillusionment should be as much of a general tactic as a massed finite and the death eaters could have been hidden.

The massively more obvious solution is just to kill Harry quickly, and moreover to not EVER offer the protagonist 60 seconds to try to save himself, no matter how interesting that sounds.

Any other general/specific tactics that LV could and should have thought of in advance? He had an entire year to plan this, and has Harry level intelligence. He should have predicted and outplayed.

View more: Next