Posts
Comments
I think it's quite in line with the attitude most porn takes.
You probably meant prefix sums instead of pairwise sums.
In any case, Bayesian reasoning is not symmetrical with respect to any given automorphism, unless the hypotheses space is.
The All-Knowing doesn't need to punish, for He can not be threatened.
Real markets mostly have it covered, because they have something close to [aggregated] utilons -- money, and so exchanges between 2 different goods rarely take place.
Also, any business can be seen as a "side-channel trade" -- the market value of one individual's time is often lower than the value they can produce in cooperation with others.
What if we suppose that wealth doesn't track merit that well, and accumulating 51% of wealth most likely signals the measurement error due to noise/randomness/luck?
And even inasmuch as it tracks capitalistic merit, it might not track other things we care about, which makes it problematic leaving all eggs in one basket.
What if we simply provide a magnetic field detector, aka compass, as an input device to our AI?
If that seems insufficient, how far before simulating full physics of a bird's body as-is would be sufficient? (It also seems that such simulation is completely outside of scope for AI, because it has nothing to do with intelligence per se).
Thank you, Duncan. I've never met you, but you seem very real and very existing to me. I don't quite share the same history as you, I think I got used to defensively ignoring what the world implied about me pretty early, but I am aiming to become a psychotherapist, and attempting to connect with how people actually are, rather than what I think they might be, seems central to my journey. Your post is an inspiration to me.
I must say I am quite taken aback by the condescending tone of your comment (suggesting that I am 15 years old etc).
But since you've got some upvotes I wonder if disagreement "with the current consensus" indeed was implied by my phrasing. In case it needs clarification, obviously, I suggest that nobody tries heroin. And even though this question seems much easier to answer, it was listed by the OP and so it would be helpful if he could first answer it himself.
UPD In case you're interested in my stance on the above substances, it's this:
-- Heroin is quite harmful.
-- Amphetamines are sometimes useful as prescription drugs, but I wouldn't recommend them otherwise.
-- I strongly encourage any adult to have LSD at least once, but with great care for the setting and risk-factors, such as relatives with schizophrenia etc.
But my stance is not the point. It's up to the OP to find his.
A couple of quick thoughts.
First, and most importantly, I think, it would be very valuable for you to try to answer these questions for yourself. Not with a goal of convincing him, but as if they had arisen in your own thoughts. Why, indeed, LSD is criminalized? What is the difference between alcohol and LSD? What is difference between coffee and amphetamines? Why not try heroin? If you are courageous enough to ponder these questions honestly, discerning what you do know from what you don't know, it will be much easier to discuss them with your son.
Second, my personal opinion is that the substances mentioned have vastly different effects, risks, and side-effects, and only in trying to rigorously outline what they are we can learn how to deal (or not to deal) with each particular one of them.
... or don't, it's a post, not a cop. I empathise with its message though.
If we look at 17!Austin and 27!Austin as two different people, then I don't see why 27!Austin would have any obligation to do anything for 17!Austin if 27!Austin doesn't want to do it.
But that's not true! Even if I don't feel obliged to 100% comply with what other people want, I certainly am affected by their desires and want to compromise. Yes, maybe it's not quite an "obligation", but I rarely experience those towards whoever anyway.
But how is it different from real life?
Apologies, I realise I might've misunderstood your comment. Do you actually mean that you shouldn't worry about being a BB because it is sort of inconsequential, not that you are sure you are not one because you've got the "next observer-moment"?
I agree with the former, but was arguing against the latter.
I wish it could work, but it doesn't. You only experience the single moment (which includes memories of the past and expectations of the future), and at this present moment you can't tell from your observations whether you are located in a proper casually-affected body, or in a BB. What's more, assuming you're a real human, your consciousness actually is not continuous, as you imply, but has intermittent gaps from 10s to 100s of ms (can't find a good citation), which you don't notice in the same way as you don't notice https://en.wikipedia.org/wiki/Saccade FWIW, you could be losing consciousness every other quantum of time, and still not notice it (and you probably do, we just can't measure it).
You obviously shouldn't care, but because it doesn't make sense to care, not due to observation.
PS Maybe you'd want to read https://en.wikipedia.org/wiki/Permutation_City by Greg Egan -- he takes the idea of simulation and discontinuity of consciousness to a lovely extreme.
I really appreciate you showing the problems with Type 1 arguments, but what I don't understand is how Type 2 arguments can seem convincing to you.
Type 2 arguments fully rely on priors, which is what you are supposed to have before any observations, in this case, before anything that happened to you in this life. The first problem is that these true priors are unchangeable, because they already existed on the first second of your life. But more importantly, I feel that arguing what the priors should be is the complete misuse of the concept. The whole point of Bayesian reasoning is that it eventually converges, regardless of the priors, and without observations you cannot possibly prefer one prior to another. If I "change my priors" after reading this post, these are not true priors anymore -- it is me updating on the evidence/observation. And so we are back to Type I.
It is as if, recognizing that you can't argue for something from observation, you are trying to find the support from outside of observation, but it doesn't really work in this case. What would you base the argument on, if everything, including your thoughts and intuitions, is either based on observation or isn't based on anything at all?
What does? On the surface it seems that plants don't have sexual selection as they don't seem to be able to affect the choice of their "partner", so they don't have the advantage of proper sexual species. But maybe I don't know enough about plants.
Because it is probably hard to isolate the applications for each behaviour from each other. "Compete with A and mate with B as much as you can" is much easier to encode than "Compete with everyone, but then maybe at some point switch to mating with however you are fighting (but be careful that they don't take advantage of it)". You get the prisoner's dilemma at the very minimum.
PS If you think about it, even in humans, who do have sexual differentiation and are capable of very complex behaviour, those behaviours are not perfectly isolated, and external aggressiveness often leaks into the family. For me it is almost out of the question that such careful delineation could exist among primitive hermaphrodites.
But in your scenario the offspring has only one "successful" parent. The best outcome for hermaphrodites would be for the "winners" to mate with each other, but then it might be unstable to switch between mating and competing behaviour between the same two creatures.
I'm surprised that the major role of sexual selection seems to be overlooked. Sexual species can speed up evolution by magnitudes of order, because the selection can happen culturally, "in the minds" (only a metaphor!). In theory, any adaptive change can happen in a single generation at once, provided that the selective behaviour is able to change unanimously. Hermaphrodites wouldn't work that well, because there is no clear distinction between the group that you are competing with and the group you are competing for, which would probably make any behavioral strategy unstable.
Well, the reason usually is "I fear it will make me look bad in the eyes of others". What next?
Your friend's case seems to be very clear cut, and the root issue is not some vague manipulation or secrecy, but the actual abuse that, I assume, continued through their interaction.
While Aella's post is very vivid in describing the horror of abuse, I don't necessarily see it in your post. You don't seem to be in a vulnerable/dependent position with respect to Carla and Dave, they don't humiliate you, don't make you doubt your own experience, don't seem to discard your feelings, and so on.
That's why if you said to me "I reserve the right to do X, if I find that you are manipulating me", I wouldn't be sure what you mean. (Even on the objective, God's eye, level, let's forget the question of how we make sure that it has indeed happened for a second).
I don't get a clear idea of what you mean by "manipulation" from your post, and I would be uncomfortable using this word as self-evident. "Making someone do something without their informed consent" seems like a reasonable attempt at a definition to me.
So the main issue, it seems, is that respecting confidentiality of some shared information may actually require a lot of effort, if it comes into conflict with the drive to behave morally. And manipulation in such cases then is bestowing this burden on a person without their informed consent, right?
Well, think about it, one-in-five is an extremely high probability! We only need 5 people to try what the OP tried, for one of them to be this successful and to write this post, and we won't hear about most of those who failed.
Yes, similarity to Pascal's wager and other religious thought is not a coincidence at all. Our existence is marked by the ultimately irreconcilable conflict: on the one hand, if you view people as ML systems, being alive is both the fundamental goal of our decision-making and the precondition for all of our world-modeling; on the other hand, we are faced with the fact that all people die.
Even if we recognize our mortality rationally, our whole subconscious is built/trained around the ever-existing subjectivity. That's why we are so often intuiting that there must be some immaculate and indestructible subject inside of us, that will not perish and can transcend space and time, as if "choosing" where to be embodied. You can see this assumption driving many thought experiments: cloning, teleportation, simulation, Boltzmann brains, cryonics, imagining what it is "to be someone else", and so on. All of them presuppose that there is something you can call you beside being you, beside the totality of your experience. Major religions call it the soul. We, rationalists, know better than to explicitly posit something so supernatural, and yet it is still hard to truly embrace the fact that if you strip away every circumstance of existence that makes you you, the perfectly abstract observer remaining is devoid of any individuality and is no more you than it is me.
What if you grab nothing from the stage but do grab the audience?
I think, if you truly grab nothing from the stage, then the audience is impersonal. My "experiencer" is exactly the same as your "experiencer", the only difference is that mine is experiencing "me", i.e., my thoughts, memories, emotions, etc.
Does it not make sense if you have a memory disease like alzheimers or concussion and somebody points as a picture and says "that is you" and you have no recollection of that, would they be wrong about it? Does it really flip on whether you feel a sense of connection to your old self?
Somebody is not wrong to use it as a social construct, but what we are discussing here, I guess, is how important it would be to me. First, it would be important to me that everyone else sees me as the descendant of that person. Second, I would still be a continuation of that person, in the ways I may not be conscious of, e.g., some past traumas, learned behaviours and so on.
Or before the disease strikes would you be wrong to worry about that person-to-be as your own welfare? Does it mean that because there is total oblivion inbetween that it doesn't happen to you?
Actually, if you think about it, we care about our future selves not necessarily because they will remember us, but because we really want to project our present selves into the future, and also because we are in the unique position to affect the lives of our future selves like no human can affect another. Both of these hold in your example.
Even before I "experience anything" when the "tabula rasa" condition is in effect the word "I" refers. Even among empty shells that have no personality or that have identical personalities "I" picks out a unique instance.
I don't think that's how psychology works. The word "I" is a concept learned with language, not something essential on its own.
To make the view extreme if you spent your life in car one could claim that "you can't leave the car" because human outside of car and an empty car is a world differently structured than human living in a car. Sure identifying with a radically transformed self might be difficult but most people think they are their future selfs (that is it is not somebody else that wakes up in their bed).
Well, it would make sense to say that it is you that left the car (or the all-encompassing simulation), as long as there is continuity of the sense of self, as long as you keep your memories and everything you've learned in your car-life. But if I'd left everything behind before I got into the simulation, I don't think it makes sense to say I am still "me".
I think, I am not getting my point across very well. The crux here is this: what is "you" but the footprint of your experience? That would include your memories, intuitions, reactions, associations, and patterns of thinking. I argue there is nothing else. If you remove that, then how "me" entering the simulation is different from "you" entering the simulation? The truly original face is devoid of any self-ness.
Let's imagine that you've spent your whole life from birth inside an MMORPG. Who is then "you", separate from the MMORPG world, who could leave it? All your dreams, hopes, desires, thoughts, intuitions, personality, and all of your sense of "self" is formed by your experience within the MMORPG. What does that leave to the "original" you? Just the structure of the neural network with which you've been born? That doesn't sound like anything essential to me. Certainly it is not something I can even be consciously aware of, how could I call it "me"? OP says "the simulators extracted and stored all of your memories", but it's an error to think that "memories" is just some data on a flash drive, if you actually remove the subtlest footprint of all your experience, then what remains?
There is a great koan which asks to "remember your original face". There are several ways to think about it, but it points to the seeming duality between the subject ("you") and the object ("the world"), that you imagine being submerged in, being an illusion, because in the end the subject is formed by the same world and is inseparable from it.
PS if you would keep your "simulated experience" though, as in "The Matrix", then the thought experiment becomes coherent, because the continuity of self is preserved, and we can really say that it is "you" that moved from one world to another, but in that case it is not clear whether I should be treating my "simulated past" differently from my "real past" -- they both have just been formative experiences that made me into who I am now.
The whole idea of a simulation you could "leave" is incoherent. It supposes that there is some part of you, which is separate from the physical/simulated world, e.g., a soul. I think there is nothing to me but the world I am experiencing, so there is no one who could leave.
A thousand times this! I haven't seen anyone pointing out what's wrong with this ritual more clearly. Exactly, we turn the celebration of individual courage into a celebration of unity/conformity, what an irony.
As a comparison, the human brain is estimated to have hundreds of trillions of synapses connecting its 100 billion neurons. In terms of numbers, Google's algorithms may soon match the human brain.
It reminds me how people first compared the amount of memory the computers had, and then their performance, as measured in FLOPS, to those of the human brain. And my intuition tells me that it is similarly misleading.
If we are talking about any sort of "optimality", we can't expect even individual humans to have these "optimal" values, much less so en masse. Of course it is futile to dream that our deus ex machina will impose those fantastic values on the world if 99% of us de facto disagree with them.
Both definitions have their issues.
"able to act on its desires unimpededly" has 2 problems. First, it is clearly describing the "agent's" (also not a well-defined category, but let's leave it at that) experience, e.g. desires, not something objective from an outside view. Second, "unimpededly" is also intrinsically vague. Is my desire to fly impeded? Is an addict's desire to quit? (If the answer is "no" to both, what would even count as impediment?) But, I guess, it is fine if we agree that "compatibilist free will" is just a feature of subjective experience.
"ability to make undetermined choices" relies on the ambiguous concept of "choice", but also would be surprisingly abundant in a truly probabilistic world. We'd have to attribute "libertarian free will" to a radioactive isotope that's "choosing" when to decay, or to any otherwise deterministic system that relies on such isotope. I don't think that agrees with intuition of those who find this concept meaningful.
You'd have to draw the line somewhere so it would have any meaning at all. What's the point in the concept if anything can be interpreted as such. What do you mean when you say "free choice" or "choice"?
I don't think it can be meaningfully defined. How could you define free choice so that a human would have it, but a complicated mechanical contraption of stones wouldn't?
I don't think computers have any more free will [free choice] than stones. Do you?
Not necessarily. Non-determinism (that future is not completely defined by the past) doesn't have anything to do with choice. A stone doesn't make choices even if future is intrinsically unpredictable. The question here is why would anyone think that humans are qualitatively different from stones.
Not necessarily. Determinism doesn't have anything to do with choice. The stone doesn't make choices regardless of determinism. The question here is why would anyone think that humans are qualitatively different from stones.
Yes! It's interesting how the concepts of agency and choice seem so natural and ingrained for us, humans, that we are often tempted to think that they describe reality deeper than they really do. We seem to see agents, preferences, goals, and utilities everywhere, but what if these concepts are not particularly relevant for the actual mechanism of decision-making even from the first-person view?
What if much of the feeling of choice and agency is actually a social adaptation, a storytelling and explanatory device that allows us to communicate and cooperate with other humans (and, perhaps more peculiarly, with our future selves)? While it feels like we are making a choice due to reasons, there are numerous experiments that point to the explanatory, after-the-fact role of reasoning in decision-making. Yes, abstract reasoning also allows us to model conceptually a great many things, but those models just serve as additional data inputs to the true hidden mechanism of decision-making.
It shouldn't be surprising then if for other minds [such as AGI], not having this primal adaptation to being modeled by others, decision-making would feel nothing like choice or agency. We already see it in our simpler AIs -- choice and reasoning mean nothing to a health diagnostic system, it simply computes, it is only for us, humans, to feel like we understand "why" it made a particular choice, we have to add an explanatory module that gives us "reasons" but is completely unnecessary for the decision-making itself!
My anecdotal experience of being a creature shows that I am very happy when I don't feel like an agent, coherent or not. The need for being an [efficient] agent only arises in the context of an adverse situation, e.g. related to survival, but agency and coherence are costly, in so many aspects. I am truly blessed when I am indifferent enough not to care about my agency or coherence.
It seems that homunculus concept is unnecessary here. You can easily talk about the experience itself, e.g. "seeing", or you can still use "I see" as a language construct while realising that you are only referring to the happening phenomenon of "seeing".
There is a difference between knowing something and experiencing it in a particular way, and the former may only very slightly nudge the latter if at all.
I can know a chair is red, but if I close my eyes, I don't see it.
I can know a chair is red, but if I put on coloured glasses, I will not see it as red.
I can know that nothing changes in reality when I take LSD, but, oh boy, does my perception change.
The real problem here is that we are not rational agents, and, what's worse, the small part of us that even resembles anything rational is not in control of our experience.
We'd like to imagine ourselves as agents and then we run into surprises like "how can I know something, but still experience it (or worse, behave!) differently".
What you are describing is data (A/B, 1/2) such that parts of the data are independent from the secret X/Y, but the whole data is not independent from the secret. That's an issue that is sort of unusual for any statistical approach, because it should be clear that only the whole leaked data should be considered.
The problem with Pearson correlation criterion is that it does not measure independence at all (even for parts of the data), but measures correlation which is just a single statistic of the two variables. It's as if you compared two distributions by comparing their means.
Let's say leaked data is X = -2, -1, 1, 2 equiprobably, and secret data is Y = X^2. Zero correlation just implies E(XY) - E(X)E(Y) = 0, which is the case, but it is clear that one can fully restore the secret from the leaked, they are not independent at all.
See more at https://en.wikipedia.org/wiki/Correlation_and_dependence#Correlation_and_independence
An important point (somewhat overlooked in the comments) is that it is not necessary to sell the humane eggs at the same price as factory eggs for this to work. You can start issuing certificates straight away, not changing anything in the distribution process!
It is even better, because it will provide incentive for competition between humane eggs producers. And also, having humane eggs still labeled as "humane", it will help them to drive factory eggs out of the market, by allowing them to have a lower profit margin. In a sense, it is similar to tax reliefs for an industry we want to stimulate, just organized in a perfectly libertarian way.
No, that's not what's wrong with Pearson's approach. Your example suffers from a different issue.