I wonder:
if you had an agent that obviously did have goals (let's say, a player in a game, whose goal is to win, and who plays the optimal strategy) could you deduce those goals from behavior alone?
Let's say you're studying the game of Connect Four, but you have no idea what constitutes "winning" or "losing." You watch enough games that you can map out a game tree. In state X of the world, a player chooses option A over other possible options, and so on. From that game tree, can you deduce that the goal of the game was to get four pieces in a row?
I don't know the answer to this question. But it seems important. If it's possible to identify, given a set of behaviors, what goal they're aimed at, then we can test behaviors (human, animal, algorithmic) for hidden goals. If it's not possible, that's very important as well; because that means that even in a simple game, where we know by construction that the players are "rational" goal-maximizing agents, we can't detect what their goals are from their behavior.
That would mean that behaviors that "seem" goal-less, programs that have no line of code representing a goal, may in fact be beh...
From that game tree, can you deduce that the goal of the game was to get four pieces in a row?
One method that would work for this example is to iterate over all possible goals in ascending complexity, and check which one would generate that game tree. How to apply this idea to humans is unclear. See here for a previous discussion.
Reductionists want to reduce things like goals and preferences to the appropriate neurons in the brain; eliminativists want to prove that humans, like the blue-minimizing robot, don't have anything of the sort until you start looking at high level abstractions.
Surely you mean that eliminativists take actions which, in their typical contexts, tend to result in proving that humans, like the blue-minimizing robot, don't have anything of the sort until you start looking at high level abstractions.
Surely you mean that there are just a bunch of atoms which, when interpreted as a human category, can be grouped together to form a being classifiable as "an eliminativist".
eliminativists want to prove that humans, like the blue-minimizing robot, don't have anything of the sort until you start looking at high level abstractions.
Just because something only exists at high levels of abstraction doesn't mean it's not real or explanatory. Surely the important question is whether humans genuinely have preferences that explain their behaviour (or at least whether a preference system can occasionally explain their behaviour - even if their behaviour is truly explained by the interaction of numerous systems) rather than how these preferences are encoded.
The information in a jpeg file that indicates a particular pixel should be red cannot be analysed down to a single bit that doesn't do anything else, that doesn't mean that there isn't a sense in which the red pixel genuinely exists. Preferences could exist and be encoded holographically in the brain. Whether you can find a specific neuron or not is completely irrelevant to their reality.
Interesting post throughout, but don't you overplay your hand a bit here?
There's nothing that looks remotely like a goal in its programming, [...]
An IF-THEN piece of code comparing a measured RGB value to a threshold value for firing the laser would look at least remotely like a goal to my mind.
More explanatory of the way people actually behave is that there's no unified preference for or against death, but rather a set of behaviors. Being in a burning building activates fleeing behavior; contemplating death from old age does not activate cryonics-buying behavior.
YES. This so much.
But if whenever I eat dinner at 6I sleep better than when eating dinner at 8, can I not say that I prefer dinner at 6 over dinner at 8? Which would be one step over saying I prefer to sleep well than not.
I think we could have a better view if we consider many preferences in action. Taking your cyonics example, maybe I prefer to live (to a certain degree), prefer to conform, and prefer to procrastinate. In the burning-building situation, the living preference is playing more or less alone, while in the cryonics situation, preferences interact somewhat like oppsite forces and then motion happens in the winning side. Maybe this is what makes preferences seem like varying?
Eliminativism is all well and good if all one wants to do is predict. However, it doesn't help answer questions like "What should I do?", or "What utility function should we give the FAI?"
The same might be said of evolutionary psychology. In which case I would respond that evolutionary psychology helped us stop thinking in a certain stupid way.
Once, we thought that men were attracted to pretty women because there was some inherent property called "beauty", or that people helped their neighbors because there was a universal Moral Law to which all minds would have access. Once it was the height of sophistication to argue whether people were truly good but corrupted by civilization, or truly evil but restrained by civilization.
Evolutionary psychology doesn't answer "What utility function should we give the FAI?", but it gives good reasons to avoid the "solution": 'just tell it to look for the Universal Moral Law accessible to all minds, and then do that.' And I think a lot of philosophy progresses by closing off all possible blind alleys until people grudgingly settle on the truth because they have no other alternative.
I am less confident in my understanding of eliminativism than of evo psych, so I am less willing to speculate on it. But since one common FAI proposal is "find out human preferences, and then do those", if it turns...
if you were in a burning building, you would try pretty hard to get out. Therefore, you must strongly dislike death and want to avoid it. But if you strongly dislike death and want to avoid it, you must be lying when you say you accept death as a natural part of life and think it's crass and selfish to try to cheat the Reaper. And therefore your reluctance to sign up for cryonics violates your own revealed preferences! You must just be trying to signal conformity or something.
I don't think this section bolsters your point much. The obvious explanation f...
A more practical example: when people discuss cryonics or anti-aging, the following argument usually comes up in one form or another: if you were in a burning building, you would try pretty hard to get out. Therefore, you must strongly dislike death and want to avoid it. But if you strongly dislike death and want to avoid it, you must be lying when you say you accept death as a natural part of life and think it's crass and selfish to try to cheat the Reaper.
nitpick: Burning to death is painful and it can happen at any stage of life. "You want to live a long life and die peacefully with dignity" can also be derived but of course it's more complicated.
So if someone stays in the haunted house despite the creaky stairwell, his preferences are revealed as rationalist?
Personally I would have run away exactly because I would not think the sound to come from a non-existent, and so harmless, ghost!
Thanks for this great sequence of posts on behaviourism and related issues.
Anyone who does not believe mental states are ontologically fundamental - ie anyone who denies the reality of something like a soul - has two choices about where to go next. They can try reducing mental states to smaller components, or they can stop talking about them entirely.
Here's what I take it you're committed to:
goals appear only when you make rough generalizations from its behavior in limited cases.
I am surprised no one brought up the usual map / territory distinction. In this case the territory is the set of observed behaviors. Humans look at the territory and with their limited processing power they produce a compressed and lossy map, here called the goal.
The goal is a useful model to talk simply about the set of behaviors, but has no existence outside the head of people discussing it.
quoted text if you were in a burning building, you would try pretty hard to get out. Therefore, you must strongly dislike death and want to avoid it. But if you strongly dislike death and want to avoid it, you must be lying when you say you accept death as a natural part of life and think it's crass and selfish to try to cheat the Reaper.
Won't it be the case that someone who tries to escape from a burning building, does so, just to avoid the pain and suffering it inflicts? It would be such a drag to be burned alive rather than a peaceful painless poison death.
Interesting that you chose the "burning building" analogy. In the fire sermon the Buddha argued that being incarnated in samsara was like being in a burning building and that the only sensible thing to do was to take steps to ensure the complete ending of the process of reincarnation in samsara ( and dying just doesn't cut it in this regard). The burning building analogy in this case is a terrible one- as we are talking about the difference between a healthy person seeking to avoid pain and disability versus the cryonics argument- which is all ab...
Excellent post!
I hope that somewhere along the way you get to the latest neuroscience suggesting that the human motivational system is composed of both model-based and model-free reinforcement mechanisms.
Keep up the good work.
Without my dealing here with the other alternatives, do you Yvain, or does any other LW reader think that it is (logically) possible that mental states COULD be ontologically fundamental?
Further, why is that possibility tied to the word "soul", which carries all sorts of irrelevant baggage?
Full disclosure: I do (subjectively) know that I experience red, and other qualia, and try to build that in to my understanding of consciousness, which I also know I experience (:-) (Note that I purposely used the word "know" and not the word "believe".)
This is an excellent post Yvain. How can I socially pressure you into posting the next one? Guilt? Threats against my own wellbeing?
I like to enforce reductionist consistency in my own brain. I like my ehtics universal and contradiction free, mainly because other people can't accuse me of being inconsistent then.
The rest, is akrasia.
Reductionists want to reduce things like goals and preferences to the appropriate neurons in the brain; eliminativists want to prove that humans, like the blue-minimizing robot, don't have anything of the sort until you start looking at high level abstractions.
I don't really see how these two philosophies contradict.
Absolutely fantastic post. Extremely clearly written, and made the blue-minimizing robot thought experiment really click for me. Can't wait for the next one.
I enjoyed reading that, in the same way that I enjoyed reading Roko's Banned Post - I don't believe it for a moment, but it stretches the mind a little. This one is much more metaphysical, and it also has an eschatological optimism that Roko's didn't. I think such optimism has no rational basis whatsoever, and in any case it means little to those unfortunates stuck in Hell to be told that Heaven is coming at the end of the computation, but unfortunately it can't come any faster because of logical incompressibility. I'm thinking of Raymond Smullyan's dialogue, in which God says that the devil is "the unfortunate length of time" that the process of "enlightenment" inevitably takes, and I think Tipler might make similar apologies for his Omega Point on occasion. All possible universes eventually reach the Omega Point (because, according to a sophistical argument of Tipler's, space-time itself is inconsistent otherwise, so it's logically impossible for this not to happen), so goodness and justice will inevitably triumph in every part of the multiverse, but in some of them it will take a really long time.
So, if I approach your essay anthropologically, it's a mix of the very new cosmology and crypto-metaphysics (of Singularities in the multiverse, of everything as computation) with a much older thought-form - and of course you know this, having mentioned Neoplatonism - but I'd go further and say that the contents of this philosophy are being partly determined by a wishful thinking, which in turn is made possible by the fundamental uncertainty about the nature of reality. In other words, all sorts of terrible things may happen and may keep happening, but if you embrace Humean skepticism about induction, you can still say, nonetheless, reality might start functioning differently at any moment, therefore I have license to hope. In that case, uncertainty about the future course of mundane events provides the epistemic license for the leap of optimism.
Here, we have the new cosmological vision, of a universe (or multiverse) dominated by the rise of superintelligence in diverse space-time locations. It hasn't happened locally yet, but it's supposed to lie ahead of us in time. Then, we have the extra ingredient of acausal interaction between these causally remote (or even causally disjoint) superintelligences, who know about each other through simulation, reasoning, or other logically and mathematically structured explorations of the multiverse. And here is where the unreasonable optimism enters. We don't know what these superintelligences choose to do, once they sound out the structure of the multiverse, but it is argued that they will come to a common, logically preordained set of values, and that these values will be good. Thus, the idea of a pre-established harmony, as in Tipler (and I think in Leibniz too, and surely many others), complete with a reason why the past and present are so unharmonious (our local singularity hasn't happened yet), and also with an extra bit of hope that's entirely new and probably doesn't make sense: maybe the evil things that already happened will be cancelled out by reversing the computation - as if something can both have happened and could nonetheless be made to have never happened. Still, I bet Spinoza never thought of that one; all he could come up with was that evil is always an absence, that all things which actually exist are good, and so there's nothing that's actually bad.
The Stoics had a tendency to equate the order of nature with a cosmic Reason that was also a cosmic Good. Possibly Bertrand Russell was the one who pointed out that this is a form of power worship: just because this is the universal order, or this is the way that things have always been, does not in itself make it good. This point can easily be carried across to the picture of superintelligences arriving at their decisional equilibrium via mutual simulation: What exactly foreordains that the resulting equilibrium deserves the name of Good? Wouldn't the concrete outcome depend on the distribution of superintelligence value systems arising in the multiverse - something we know nothing about - and on the resources that each superintelligence brings to the table of acausal trade and negotiation? It's intriguing that even when confronted by such a bizarrely novel world-concept, the human mind is nonetheless capable, not only of interpreting it in a way originating from cultures which didn't even know that the sun is a star, but of finding a way to affirm the resulting cosmology as good and as predestined to be so.
I have mentioned Russell's reason for scorning the Stoic equation of the cosmic order with the cosmic good (it's just worship of overwhelming force), but I will admit that, from an elemental perspective which values personal survival (and perhaps the personal gains that can come from siding with power), it does make sense to ask oneself what the values of the hypothetical future super-AI might be. That is, even if one scorns the beatific cyber-vision as wishful thinking, one might agree that a future super-purge of the Earth, conducted according to the super-AI's value system, is a possibility, and attempt to shape oneself so as to escape it. But as we know, that might require shaping oneself to be a thin loop, a few centimeters long, optimized for the purpose of holding together several sheets of paper.
I acknowledge your points about not equating Goodness with Power, which is probably the failure mode of lusting for reflective consistency. (The lines of reasoning I go through in that link are pretty often missed by people who think they understand the nature of direction of morality, I think.) Maybe I should explicitly note that I was not at all describing my own beliefs, just trying to come up with a modern rendition of old-as-dirt Platonistic religionesque ideas. (Taoism is admirable in being more 'complete' and human-useful than the Big Good Metaphysi...
Anyone who does not believe mental states are ontologically fundamental - ie anyone who denies the reality of something like a soul - has two choices about where to go next. They can try reducing mental states to smaller components, or they can stop talking about them entirely.
In a utility-maximizing AI, mental states can be reduced to smaller components. The AI will have goals, and those goals, upon closer examination, will be lines in a computer program.
But in the blue-minimizing robot, its "goal" isn't even a line in its program. There's nothing that looks remotely like a goal in its programming, and goals appear only when you make rough generalizations from its behavior in limited cases.
Philosophers are still very much arguing about whether this applies to humans; the two schools call themselves reductionists and eliminativists (with a third school of wishy-washy half-and-half people calling themselves revisionists). Reductionists want to reduce things like goals and preferences to the appropriate neurons in the brain; eliminativists want to prove that humans, like the blue-minimizing robot, don't have anything of the sort until you start looking at high level abstractions.
I took a similar tack asking ksvanhorn's question in yesterday's post - how can you get a more accurate picture of what your true preferences are? I said:
A more practical example: when people discuss cryonics or anti-aging, the following argument usually comes up in one form or another: if you were in a burning building, you would try pretty hard to get out. Therefore, you must strongly dislike death and want to avoid it. But if you strongly dislike death and want to avoid it, you must be lying when you say you accept death as a natural part of life and think it's crass and selfish to try to cheat the Reaper. And therefore your reluctance to sign up for cryonics violates your own revealed preferences! You must just be trying to signal conformity or something.
The problem is that not signing up for cryonics is also a "revealed preference". "You wouldn't sign up for cryonics, which means you don't really fear death so much, so why bother running from a burning building?" is an equally good argument, although no one except maybe Marcus Aurelius would take it seriously.
Both these arguments assume that somewhere, deep down, there's a utility function with a single term for "death" in it, and all decisions just call upon this particular level of death or anti-death preference.
More explanatory of the way people actually behave is that there's no unified preference for or against death, but rather a set of behaviors. Being in a burning building activates fleeing behavior; contemplating death from old age does not activate cryonics-buying behavior. People guess at their opinions about death by analyzing these behaviors, usually with a bit of signalling thrown in. If they desire consistency - and most people do - maybe they'll change some of their other behaviors to conform to their hypothesized opinion.
One more example. I've previously brought up the case of a rationalist who knows there's no such thing as ghosts, but is still uncomfortable in a haunted house. So does he believe in ghosts or not? If you insist on there being a variable somewhere in his head marked $belief_in_ghosts = (0,1) then it's going to be pretty mysterious when that variable looks like zero when he's talking to the Skeptics Association, and one when he's running away from a creaky staircase at midnight.
But it's not at all mysterious that the thought "I don't believe in ghosts" gets reinforced because it makes him feel intelligent and modern, and staying around a creaky staircase at midnight gets punished because it makes him afraid.
Behaviorism was one of the first and most successful eliminationist theories. I've so far ignored the most modern and exciting eliminationist theory, connectionism, because it involves a lot of math and is very hard to process on an intuitive level. In the next post, I want to try to explain the very basics of connectionism, why it's so exciting, and why it helps justify discussion of behaviorist principles.