From: Keith Henson (email@example.com)
Date: Wed Feb 18 2004 - 22:53:37 MST
At 10:37 PM 15/02/04 -0500, you wrote:
>Keith Henson wrote:
>>>The point is that you can perform all learning necessary to the task of
>>>transforming everything in sight into paperclips, and you won't have
>>>conflicts with distant parts of yourself that also want to transform
>>>everything into paperclips - the target is constant, only the aim gets updated.
>>And the weapons, and the gravity field, and if it starts thinking about
>>what it is doing and what paper clips are used for, it might switch from
>>metal to plastic or branch out into report covers (which do a better job)
>>and then reconsider the whole business of sticking papers together and
>>start making magnetic media.
>What does it matter so long as there are paperclips?
>Seriously, what *does* it matter from the perspective of a mind that only
>wants paperclips? If you yourself want something besides paperclips, you
>should not build a paperclip optimization process, of course.
Agreed. But "mind" is by definition more than an automaton. I.e., we
already have paperclip making machines. The goals of giving a mind a
single focused nature *and* great intellect would seem to be in conflict.
>>>Hm... I infer that you're thinking of some algorithm, such as
>>>reinforcement on neural nets, that doesn't cleanly separate model
>>>information and utility computation.
>>Even if you cleanly split out utility computation, widely separated AIs
>>are going to be working off rather different data bases.
>>Take shifting a galaxy to avoid the worst consequences of collisions
>>(whatever they are). That's an obvious project for a friendly and very
>>patient AI. Say galaxy A needs to go right or left direction and galaxy
>>B needs to go left or right depending on what A does to modify the
>>collision. If the AIs figure this out when they are separated by several
>>million light years, they are going to have a heck of a time deciding
>>which way each should cause their local galaxy to dodge. If they both
>>decide the same way, you are going to get one of those sidewalk episodes
>>of people dodging into each other's path--with really lamentable results
>>for anybody nearby if the black holes merge.
>If you anticipate this problem in advance, you can keep a simple reference
>mind on offline storage somewhere, and some set of agreed-on protocols for
>reducing your local data to the subset of the local data that would be
>visible to a distant self. Both copies of yourself feed the reference
>mind identical copies of the intersection of the data that would be known
>to both entities. The reference mind then outputs a set of coordinated
>high-level strategies on the level where coordination is necessary. The
>rest is up to the local minds and they can use full knowledge in
>In general, the ability to carry out optimal plans with multiple actions,
>whether simultaneous spatially distributed actions or temporally
>distributed local actions, depends on your ability to reliably predict
>spatially or temporally distant actions. The solution I gave above is an
>extreme case of the answer, "in thinking through coordinated plans, don't
>use data your other self can't access". This answer is not necessarily
>optimal, but it's simple. A more complex answer would involve optimizing
>over probability distributions for the distant mind's action. The more
>important it is to be perfectly coordinated, the more unshared information
>you should throw away in order to be predictable.
Twin brothers living far apart and out of communication each constructs the
first (or second) automobiles. They complete their lethally fast toys on
the same hour of the same day and each decides to visit the other. In
spite of these being the first two cars, there is a well paved, two vehicle
wide road between the homes of the two brothers. Each figures out that
driving in the center would be a disaster if someone else also had a
car. But which side of the road do they pick?
(A random pick results in 50% chance of killing both of them in a fatal
head on collision.)
>>To the extent humans share goals it is because humans share genes. Males
>>in particular are optimize to act and to take risks for others on the
>>basis of the average relationship in a tribe a few hundred thousand years
>>ago (averaging to something like second cousin).
>Sometimes humans share goals, not because they have high relatedness to
>one another, but because humans share the genes that construct the goals
Or memes are selected in the environment of brains which seem to give a
group a common goal. Cult memes for example.
>and the goals are cognitively implemented in non-deictic form (the goal
>template doesn't use the "this" variable).
> For example, humans like particular kinds of environments, so if you
> were to propose a workable way of transforming Toronto into the tree-city
> of Lothlorien, there'd be widely distributed support for that proposal
> not because everyone in Toronto is related to you, but because the parts
> of our brains that process the pretty flowers (signs of fertile
> territory) are constructed by species-typical genes. Shared utility
> functions exist because of shared genes, but not necessarily because of
> Hamiltonian relatedness.
A lot of this already happens. Our liking for parks of mowed grass and
nearby trees was obtained honestly by our remote ancestors.
>Likewise, you can get selection pressures derived from iterated Prisoner's
>Dilemma between not necessarily related partners, and selection pressures
>on more complex social interactions if language is around. If you had an
>evolved intelligent species whose spawning process scrambled zygotes
>spatially before they grew up, so that they weren't related to nearby
>individuals, I'd still expect them to evolve social coordination
>mechanisms in the process of evolving intelligence. We behave honorably
>toward unrelated individuals.
That's true, but I really wonder if it didn't take close proximity to
relatives to shape up the ability to behave honorably toward
unrelateds? On the other hand, octopi spawn that way. Anyone have a
pointer to them acting honorably toward each other?
The point is perhaps moot unless someone uses an evolutionary process to
>>I have a real problem of part of my brain being subjective months our of
>>sync. When you have to communicate, even with your twin brother, via
>>sailing ship you might have the same interest and goals but you darn sure
>>are going to be different individuals.
>The human side of this is one issue; making lots of paperclips, or
>creating a stable FAI, is another. Obviously you can't have brain lobes
>millions of ticks distant from each other and remain a classical human.
>I'm just saying it doesn't obviously introduce insoluble stability
>problems for an FAI.
One reason this concerns me is personal boosted intelligence. If we are
going to deal with AIs as something more than pets, we are probably going
to need enhancement. But talking to an AI that was spread out over light
minutes might be a bit disconcerting.
A point I understand but slightly question is the assumption that there
well be only once and the others will be clones. I can see this if the
takeoff and spreading out is extremely fast, but it if is not you have the
potential for more than one AI. Does more than one creat a problem?
This archive was generated by hypermail 2.1.5 : Wed May 22 2013 - 04:00:38 MDT