Comment author: CoffeeStain 07 September 2013 09:29:41PM *  12 points [-]

Instead of friendliness, could we not code, solve, or at the very least seed boxedness?

It is clear that any AI strong enough to solve friendliness would already be using that power in unpredictably dangerous ways, in order to provide the computational power to solve it. But is it clear that this amount of computational power could not fit within, say, a one kilometer-cube box outside the campus of MIT?

Boxedness is obviously a hard problem, but it seems to me at least as easy as metaethical friendliness. The ability to modify a wide range of complex environments seems instrumental in an evolution into superintelligence, but it's not obvious that this necessitates the modification of environments outside the box. Being able to globally optimize the universe for intelligence involves fewer (zero) constraints than would exist with a boxedness seed, but the only question is whether or not this constraint is so constricting as to preclude superintelligence, which it's not clear to me that it is.

It seems to me that there is value in finding the minimally-restrictive safety-seed in AGI research. If any restriction removes some non-negligible ability to globally optimize for intelligence, the AIs of FAI researchers will be necessarily at a disadvantage to all other AGIs in production. And having more flexible restrictions increases the chance than any given research group will apply the restriction in their own research.

If we believe that there is a large chance that all of our efforts at friendliness will be futile, and that the world will create a dominant UFAI despite our pleas, then we should be adopting a consequentialist attitude toward our FAI efforts. If our goal is to make sure that an imprudent AI research team feels as much intellectual guilt as possible over not listening to our risk-safety pleas, we should be as restrictive as possible. If our goal is to inch the likelihood that an imprudent AI team creates a dominant UFAI, we might work to place our pleas at the intersection of restrictive, communicable, and simple.

Comment author: Eugene 11 October 2013 07:50:53PM *  0 points [-]

A slightly bigger "large risk" than Pentashagon puts forward is that a provably boxed UFAI could indifferently give us information that results in yet another UFAI, just as unpredictable as itself (statistically speaking, it's going to give us more unhelpful information than helpful, as Robb point out). Keep in mind I'm extrapolating here. At first you'd just be asking for mundane things like better transportation, cures for diseases, etc. If the UFAI's mind is strange enough, and we're lucky enough, then some of these things result in beneficial outcomes, politically motivating humans to continue asking it for things. Eventually we're going to escalate to asking for a better AI, at which point we'll get a crap-shoot.

An even bigger risk than that -though - is that if it's especially Unfriendly, it may even do this intentionally, going so far as to pretend it's friendly while bestowing us with data to make an AI even more Unfriendly AI than itself. So what do we do, box that AI as well, when it could potentially be even more devious than the one that already convinced us to make this one? Is it just boxes, all the way down? (spoilers: it isn't, because we shouldn't be taking any advice from boxed AIs in the first place)

The only use of a boxed AI is to verify that, yes, the programming path you went down is the wrong one, and resulted in an AI that was indifferent to our existence (and therefore has no incentive to hide its motives from us). Any positive outcome would be no better than an outcome where the AI was specifically Evil, because if we can't tell the difference in the code prior to turning it on, we certainly wouldn't be able to tell the difference afterward.

Comment author: ciphergoth 02 September 2013 05:50:17PM 0 points [-]

I can't find it on Google now, but ISTR that the meaning of "abet" you give here is an urban legend; "abet" just means "aid" and was only put in to make it sound more grand.

Comment author: Eugene 07 September 2013 11:09:46PM *  3 points [-]

It is not an urban legend. From etymonline:

from a- "to" + beter "to bait," from a Germanic source, perhaps Low Franconian betan "incite," or Old Norse beita "cause to bite"

Comment author: linkhyrule5 29 August 2013 12:38:25AM 6 points [-]

They see you as small and helpless, They see you as just a child...

Wow. I expect that from Harry at this point, but this just rubbed in the fact that the eleven-year-old protagonists are very much more heroic than most of the adults.

So, here's a question: Aside from uniting Malfoy and Bones, and in general every House in Hogwarts, and the Boy-Who-Lived on top of that in to a single anti-childkilling power bloc, what else is going on here?

The first thing that comes to mind is that this is probably part of Quirrell's plot to set up Harry as Light Lord...

Comment author: Eugene 07 September 2013 10:56:23PM *  0 points [-]

"The first thing that comes to mind is that this is probably part of Quirrell's plot to set up Harry as Light Lord..."

If it's as patently ridiculous as his plot to invent a fake Dark Lord who publicly reveals himself and challenges Harry to a fake public duel where he casts a fake Avada Kedavra that fake-backfires just so Harry can spend summer vacation at home, then I sure hope not.

Comment author: Viliam_Bur 31 August 2013 07:46:33PM *  0 points [-]

Perhaps the reason is that a rationalist wouldn't waste time. A superior mind does not need 7 years to conquer the world with magic. It just needs to find the ways to recursively self-improve and then... FOOM!

Comment author: Eugene 07 September 2013 10:45:08PM 4 points [-]

That fine, except a perfect rationalist doesn't exist in a bubble, nor does Harry. Much of what's making the story feel rushed isn't Harry's actions, but rather the speed at which those actions propagate among people who are not rational actors.

Harry is not an above-human-intelligence AI with direct access to his source code. Therefore he cannot "FOOM", therefore he's stuck with a world that is still largely outside his ability to control, no matter how rational he is.

Comment author: PhilGoetz 04 September 2013 09:24:33PM *  9 points [-]

Draco seems insufficiently clever--he's gotten a lot of character development, and could be an agent in the story, but so far he just reacts. (He's already too clever for a 10-year old, but Harry is far too clever for a 10-year-old, so fair's fair.)

I can imagine an alternate HPMoR in which instead of Hermione, Harry had died--preferably later in the story, say at a false climax in the middle of a full-scale war--and Hermione and Draco had to step up to being the protagonists. Possibly I'm just getting tired of Harry's perfection, but seeing them develop towards what Harry is now seems to have more potential drama than seeing Harry develop anywhere (that can be portrayed in fiction and understood by a wide audience) from where he is already.

Comment author: Eugene 07 September 2013 10:04:27PM *  0 points [-]

Wrong thread

Comment author: IsaacWheeland 02 July 2013 02:58:21AM 8 points [-]

I emailed back, and she elaborated:

"Oh, no, my issue is not with the fact that Eliezer killed a female character for Harry's motivation. Like I said, I like character death. I like character death used to put other characters through an emotional rollercoaster. And when people complain that X is sexist because a female character got fridged, that annoys me because while the trend is an issue, there is nothing wrong or sexist with an individual instance of a character who happens to be female dying for a character who happens to be male. It actually kind of surprised me on a meta level that I was so mad - I have never been annoyed by an individual instance of fridging before.

But the issue here is with the context in this particular instance. Your argument that it had to happen this way is flawed, because it assumes the story prior to the exact point of Hermione's death was fixed and out of Eliezer's hands. I don't have a problem with Eliezer killing Hermione, in itself - but if he was going to, he should have either not given her this character arc in the first place or completed the arc first in a way that gives at least some vague kind of closure. He could also have killed Neville if he wanted to - he'd just have needed to develop Harry's relationship with Neville in such a way that it would make sense as a motivator, instead of (or along with) his relationship with Hermione. And it's not as if he suddenly realized here after writing the story up to this point that he needed to kill Hermione in order for it to work out - the trigger warnings page has noted that the next chapter with a trigger warning would be called "The Bystander Effect" (he notes specifically on chapter 88 that the original title was "Bystander Apathy", clearly as a way of alerting those who have been watching out for the next triggery chapter that this is it) since August 2010. This was planned. He knew exactly how he was going to kill Hermione, and he had all the time in the world to plan out a way for it to go that wouldn't involve aborting a potentially interesting arc and making Hermione The Character Who Could Never Step Out Of Harry's Shadow And Then Died.

If you've watched Game of Thrones (or read A Song of Ice and Fire, assuming this bit is more or less the same as the show), it also shockingly kills off main characters a lot, but while it is shocking and unexpected, it is not unsatisfying like this, because the characters who are killed, while they had personalities and plans and development, didn't have arcs going much of anywhere in particular at the moment - the story wouldn't have been any better with them remaining alive than dead at that point. I feel this is very distinctly not the case for Hermione in MoR. She had interesting stuff left to do. She had been written with uncomfortable overtones (taking a canon character one of whose main qualities was being smart and repeatedly making her fail where Harry succeeds because he's more rational), but the story suggested her arc was about her discovering her own way towards not having to be in Harry's shadow anymore, which would have fixed it. By aborting the arc, all that's left is those uncomfortable overtones of glorifying Harry and belittling Hermione - a character who happens to have canonically been the intelligent one, who taught millions of girls that being smart could be pretty badass.

And now MoR's only remaining vaguely developed female character is McGonagall, who, while fun, is also repeatedly emphasized as being markedly irrational (in this very chapter, even). From a source material that did at least reasonably well with female characters, after a story that seemed to be heading towards also doing at least okay on that front, Eliezer ended up with a story about how much better a boy is than almost everybody else, where all of the rare exceptions are male and the one female character who could have held her own gets fridged before anything comes of it. As a canon-Hermione fan, I feel pretty damn slapped in the face, especially when this comes straight after chapter 87 (which also irritated me a lot by making Hermione preoccupied with her tiresome ~unrequited love~ for Harry when she's just been framed for murder and should have way better things to think about). It did not have to be this way.

And, like I said, even if he absolutely had to develop an arc and make it look like Hermione was going to get something done in her life only to have it not happen to make her death more shocking, he could still have done it without the pathetic damseling, and I might have been able to let it slide. Couldn't this have been from Hermione's point of view up until when Harry arrives at the scene? (He could still have done Harry's viewpoint, too.) Couldn't we have actually seen her attempts to fight it off (which Eliezer could at least have attempted to make somewhat awesome)? Couldn't Harry have arrived sometime before Hermione became a helpless immobile McGuffin, and seen her holding her own at least somewhat? (If chapter 90 is Hermione's fight with the troll from her point of view and it's awesome and has some kind of closure to her character, or Hermione gets brought back in some manner and gets some closure afterwards, I'll be reasonably content.)

The end result, as I also mentioned, is that I don't even find it heartwrenching, because I'm too busy being angry at the context to be immersed in the story at all anymore at that point. I couldn't even concentrate when reading the whole last bit because all I wanted to do was start to type a rant into the comment box. If I had managed to be emotionally impacted by it, I'd probably also be more inclined to forgive it for the sake of good storytelling, but exactly because the context was so maddening, I couldn't. I was more saddened by the death of Mrs. Norris than Hermione.

I posted my review on FFN, and he says he reads all reviews, so he's probably read it already, but you can post it if you want. (Probably better include my elaborations here.)"

Comment author: Eugene 07 September 2013 09:42:28PM 1 point [-]

I can't help but observe that even if Hermione had been male, and just Harry's friend - even if we take out all notions of sexism or relationship dynamics from this problem - killing him off is still not really the best solution. This was a character who was growing, who was admittedly more interesting than Harry, and who was on a path that could've potentially put this character at or even above Harry's level of rational thinking. But now we're just left with Harry again, and it feels like settling for second-best.

Perhaps later chapters will convince me otherwise, but for now I am suspicious that the direction this story is going is not the best direction for this story.

Comment author: Eugene 07 September 2013 09:03:10PM *  1 point [-]

Funny thing about this chapter: up until now, I was growing fairly convinced that if any major character was going to die early, the most logical choice would be Harry. His character arc was plateauing while Hermione's was growing ever larger, many loose ends about himself were being tied up, and new ordeals were arising which propped up either-or-both Draco and Hermione as potential candidates for being the true protagonist(s) of the story. Unfortunately, the events of this chapter have at least given an appearance of permanently closing that path forward. I'm afraid this leaves us with - I claim at my own risk - a more predictable story than I was anticipating.

Granted, I don't mean to claim the author has shot himself in his own foot. Although I will comment that he appears to be doing everything in his power to try. Given two stories with happy endings - one where Hermione dies early and one where Harry dies early - the second story is clearly the most interesting challenge, presents the more exciting of the two puzzles, and is much harder to predict for the reader.

But to be fair, that doesn't mean the first isn't also worth reading. After all, I recognize that the primary goal of the story is to advance lessons about using rationality, which is far easier to accomplish when your main character is a rational actor already, rather than someone on the road to becoming a rational actor. As such, it may have simply been outside of Eliezer's skill-set to effectively or confidently continue imparting lessons while impaired with the further challenge of working with developing - rather than developed - rationalists as the main characters driving the story onward. Even if this were not the case and Eliezer does have the means for crafting that story, it still would be reasonable to predict that such a challenge would take the story much, much longer to write than perhaps the author was willing to consider acceptable. A disappointing decision, no doubt, but we all have to manage our time.

Still, what a fascinating challenge that would have been...

Comment author: tgb 03 July 2013 02:34:07AM 1 point [-]

HPMOR prediction of low confidence: The Weasleys' use of the deligitor prodi spell gives Harry a good way to access high-level magical artifacts. If the Hat shouts 'Gryffindor!' to the Weasleys, might it not shout 'Ravenclaw' to Harry? There are no magical items from canon other than maybe the stone that I would expect Harry to want more than the Diadem. And surely Harry has the proper motive to use the Hat, regardless of whether it would require a Ravenclaw or a Gryffindor motive to get the diadem.

Comment author: Eugene 07 September 2013 08:40:41PM *  0 points [-]

There's a problem with that. The Hat expressly forbade Harry to ever wear it again, since that leads to troubling Sentience issues. While that might potentially make it vastly more powerful in his hands than in others, I have serious doubts that it would actually come if called that particular way.

Comment author: JoshuaZ 02 July 2013 02:02:07AM 4 points [-]

If doing magic for more time makes one stronger (which seems to be a hypothesis taken seriously in HPMR), then it is possible that as one gets more powerful, the increased power can compensate for the incorrect pronunciation. In fact, this also may explain to some extent how less powerful witches and wizards can't cast some spells. In some cases it may be that the orally transmitted version of the spell is not quite right, but that doesn't matter as much for the more powerful spellcasters. A problem with this hypothesis is that one would then expect there to be weak spells which could only be cast by powerful mages and we haven't seen any indication of that.

Comment author: Eugene 07 September 2013 08:23:45PM 2 points [-]

Point against: Professor Whatsisname, the presumably quite-powerful dueling legend, learned/developed "Stuporfy", which is intentionally meant to sound almost exactly like "Stupify". If powerful wizards get a pass on their pronunciation, how is it that a powerful wizard can effectively differentiate those two similar spells when casting?

Comment author: matabele 30 July 2012 12:28:36PM -2 points [-]

Votes can not be counted more than once, and every vote counts (according to the voter.) As all voters have an equal opportunity to withhold or spend votes - how can this be unfair?

In current systems, a minority voter may never be offered a candidate worth a vote - all such votes don't count (according to the voter.) This is clearly unfair, and has only an appearance of proportional representation.

With the right maneuvering among a well-organized block of voters, I could imagine a situation where the system becomes a perpetual minority rule.

And this does not happen now?

This is likely the reason for low turn outs in many elections - the voters simply do not care.

Comment author: Eugene 17 August 2012 03:41:17AM *  1 point [-]

That's just the problem. It does happen now, in a system where everyone is throttled at only one vote to spend per election. In a system where you can withhold that vote till another election, increasing the power of your vote over time, it only exacerbates this behavior.

Is the better fairness on a micro level worth the trade-off of lesser fairness on a macro level?

View more: Prev | Next