Consider three kinds of ancestry trees: 1) souls of some odd human mothers, 2) ems and their copies, and 3) splitting quantum worlds. In each kind of tree, agents can ask themselves, “Which future version of me will I become?”
One clue to the mystery of the Born rule is that evolution of a wavefunction only preserves its 2-norm, so under a probability rule with any exponent not equal to 2, you'd be able to solve problems in PP in polynomial time using post-selection. See Scott Aaronson on Born Probabilities where I tried to explain this result and how it might solve the mystery, or see Scott's original paper.
I know that's not exactly the point of your essay, but this point seems not so clear to me:
"(The agent) will become all descendant branches in that tree."
Assume we have on t1 an agent A, and from this agent at t2 two descendant agents B1 and B2. Now if A becomes B1 and B2, A is personally identical to B1, and to B2. But with transitivity and symmetry this would mean that B1 is identical to B2! Which is false, I presume.
Now if we say instead that A, B1, and B2 are not agents, but just temporal parts of agents, we would have two agents: AB1 and AB2. And those two are not identical. Problem solved? But then these two agents would partially overlap, namely in respect to A at t1! At t1, A could truthfully say: "I am two agents!" But this seems no less absurd than the problem above.
This is a problem that's bothered me for some time. It seems as though either some branches end up with implausibly low probablilites or some form of retrocausality in assigning branch weightings is needed.
For instance--if one uses the consistent 1/2 weighting for the outcome of every branch event so popular with vulgar metaphysicians, then one can trivially arrange a series of just 30 duplications such that the final clone gets a weighting of 1/2^30, so you needn't worry much before the fact if said clone of yours is slated to be thrown into a volcano, as your odds of being killed by lightning or even by an asteroid are far greater than that of your being that unfortunate clone. If OTOH each clone gets an equal weight of 1/31, irrespective of its history, then you have cause to be gravely concerned.
But as you note above each endpoint getting equal weight necessarily introduces retrocausality of some sort, and raises very weird "quantum suicide" issues in the quantum multiverse case, where each branch can presumably get an astronomical number of extra "splittings" for each additional second it continues.
In the multiverse case there's also potentially the alarmingly weird scenario of different conscious observers having a different "prog prog" for the same future multiverse branch. If we allow arbitrary rules to be considered it can get very weird indeed, with differerent observers becoming highly "concentrated" in disperate mulitverse branches.
Imagine three observers, Fred, Mary, and Martha. All three are seated in front of a partly-silvered sheet of glass, mirror, with a special light source mounted behind them. When a photon originating from this special light source strikes the glass, it splits the universe in such a way that Fred has a 1% chance of being on the branch where the photon reflects off the glass and a 99% chance of being on the branch where it passes through it, Mary has a 50% chance of being on the branch where it reflects and a 50% chance of being on the branch where it passes through, and Martha has a 99% chance of being on branch where it reflects and a 1% chance of being on the branch where it passes through.
When Fred looks at the glass he sees a window, when Mary looks at the glass she sees a two-way mirror, and when Martha looks at the glass she sees a conventional mirror. When Fred asks Mary and Martha what they see, they confirm that it is indeed a window. When Mary asks the others, they confirm that the glass a two-way mirror. When Martha asks the others, they confirm that it is an ordinary mirror. There is no contradiction here--it's just that noone is aware that they are talking to astronomically unlikely "probability ghost" versions of their comrades. And of course, these "probability ghosts" must still exist even in a naive "normal" quantum multiverse.
There is the complication of just what constitutes the "you" that the prog prob is tracking. Base physical reality presumably is just a bunch of interacting quantum fields with no clear dividing line between one individual and another. If our brains our connected by a high-bandiwidth link are we one conscious individual or two for the purposes of assigning branch probablities? What if we just exchange a few words? What if we just gesture at each other? What if we communicate using smoke signals? Perhaps reasoning along these lines could somehow rule out the odder "probability ghost" scenarios in a manner similar to Galileo's thought experiment about falling objects.
"On the other hand as human beings we feel a duty to continue our journey. We feel a pressure to continue nation states,human species, intelligence etc."
Speak for yourself. These are parasitic memes that should never have existed and that have no valuable function for us cooperative individuals. You can count self-sacrifice for gods, nature, future generations or life in general among them also. Whenever someone calls for that, they are trying to parasitize you. The only exception on the list is probably the nation state which can be seen partly as a pragmatic cooperative of (many) individuals, e.g. to organize mutually beneficial military defense against other (aggressive) nation states. But that is highly conditional; nation states also do many things that don't benefit their citizens. Plus it needs to come with time discounting, because all individuals die in time.
Generally speaking, "moral" memes that call for self-sacrifice of individuals for the benefit of abstract systems or non-reciprocators should be thrown into the dustbin of social acceptability.
That's the first horn of the trilemma, in the link:
"Buy a ticket. Suspend your computer program just before the lottery drawing - which should of course be a quantum lottery, so that every ticket wins somewhere. Program your computational environment to, if you win, make a trillion copies of yourself, and wake them up for ten seconds, long enough to experience winning the lottery. Then suspend the programs, merge them again, and start the result. If you don't win the lottery, then just wake up automatically."
"You could say, "I have no problem with the idea that no one else could see you exerting your anthropic psychic powers, and I have no problem with the idea that different people can send different portions of their subjective futures into different realities."
I find myself somewhat reluctant to bite that bullet, personally."
Thanks for pointing out the typo; fixed now. In my model, large worlds see almost no distortion, small worlds are totally mangled, and the transition region is a small % of the total log range. Yes, tis frustrating.
Last sentence: "progeny prog" -> "progeny prob"
In your growth-drift-diffusion-absorption model which reproduces the simple Born rule inexactly, what is the explanation for our finding ourselves in a world in which there is no apparent historical discrepancy from exact compliance to that rule?
Aside: It must be frustrating to have made specific predictions which may be confirmed or refuted by theoretical analysis, and to not be able to get sufficient expert attention to move that analysis forward.
I don't understand your question. It isn't crazy to be concerned about large upcoming changes, but there's a wide range between "terrified and revolt" and "feel relieved". We'll most likely muddle through in the middle.
As human beings we know that suffering exists. Many philosophersand philosophies have been building their ideas that it would have better never have been.
On the other hand as human beings we feel a duty to continue our journey. We feel a pressure to continue nation states,human species, intelligence etc. As an analogy we could say that we are carrying a torchlight that has been bestowed to us from previous generations and we should bestow it to next generations. It is a torchlight of the most intelligent species known.
What do you think how people would feel if we had an inception of strong AIs, EMs or transhumanistic life forms?
sqenario one:-In buddhism one basic teaching is impermanence. Even if suffering exists, impermanence gives us hope. In this way we could feel relieved. The weight of the consciousness would be of from our chest. Our suffering, status competion etc. would be impermanent and we could throw the torchlight of intelligence to some transhumanistic life forms. Now it would be transhumanistic life forms duty to continue the journey and we would have done our share.
sqenario two:-people would feel terrified and revolt
Which one of these two sqenarios is closer to your thinking?
That is more a clue to what the rule must be than to how that rule actually comes about mechanically.
One clue to the mystery of the Born rule is that evolution of a wavefunction only preserves its 2-norm, so under a probability rule with any exponent not equal to 2, you'd be able to solve problems in PP in polynomial time using post-selection. See Scott Aaronson on Born Probabilities where I tried to explain this result and how it might solve the mystery, or see Scott's original paper.
This would mean no agent exists over time. At every instance in time you would have a different agent.
"will become" is not the same as "identical to".
I know that's not exactly the point of your essay, but this point seems not so clear to me:
"(The agent) will become all descendant branches in that tree."
Assume we have on t1 an agent A, and from this agent at t2 two descendant agents B1 and B2. Now if A becomes B1 and B2, A is personally identical to B1, and to B2. But with transitivity and symmetry this would mean that B1 is identical to B2! Which is false, I presume.
Now if we say instead that A, B1, and B2 are not agents, but just temporal parts of agents, we would have two agents: AB1 and AB2. And those two are not identical. Problem solved? But then these two agents would partially overlap, namely in respect to A at t1! At t1, A could truthfully say: "I am two agents!" But this seems no less absurd than the problem above.
Er, "sheet of glass, mirror, with" --> "sheet of glass with"
This is a problem that's bothered me for some time. It seems as though either some branches end up with implausibly low probablilites or some form of retrocausality in assigning branch weightings is needed.
For instance--if one uses the consistent 1/2 weighting for the outcome of every branch event so popular with vulgar metaphysicians, then one can trivially arrange a series of just 30 duplications such that the final clone gets a weighting of 1/2^30, so you needn't worry much before the fact if said clone of yours is slated to be thrown into a volcano, as your odds of being killed by lightning or even by an asteroid are far greater than that of your being that unfortunate clone. If OTOH each clone gets an equal weight of 1/31, irrespective of its history, then you have cause to be gravely concerned.
But as you note above each endpoint getting equal weight necessarily introduces retrocausality of some sort, and raises very weird "quantum suicide" issues in the quantum multiverse case, where each branch can presumably get an astronomical number of extra "splittings" for each additional second it continues.
In the multiverse case there's also potentially the alarmingly weird scenario of different conscious observers having a different "prog prog" for the same future multiverse branch. If we allow arbitrary rules to be considered it can get very weird indeed, with differerent observers becoming highly "concentrated" in disperate mulitverse branches.
Imagine three observers, Fred, Mary, and Martha. All three are seated in front of a partly-silvered sheet of glass, mirror, with a special light source mounted behind them. When a photon originating from this special light source strikes the glass, it splits the universe in such a way that Fred has a 1% chance of being on the branch where the photon reflects off the glass and a 99% chance of being on the branch where it passes through it, Mary has a 50% chance of being on the branch where it reflects and a 50% chance of being on the branch where it passes through, and Martha has a 99% chance of being on branch where it reflects and a 1% chance of being on the branch where it passes through.
When Fred looks at the glass he sees a window, when Mary looks at the glass she sees a two-way mirror, and when Martha looks at the glass she sees a conventional mirror. When Fred asks Mary and Martha what they see, they confirm that it is indeed a window. When Mary asks the others, they confirm that the glass a two-way mirror. When Martha asks the others, they confirm that it is an ordinary mirror. There is no contradiction here--it's just that noone is aware that they are talking to astronomically unlikely "probability ghost" versions of their comrades. And of course, these "probability ghosts" must still exist even in a naive "normal" quantum multiverse.
There is the complication of just what constitutes the "you" that the prog prob is tracking. Base physical reality presumably is just a bunch of interacting quantum fields with no clear dividing line between one individual and another. If our brains our connected by a high-bandiwidth link are we one conscious individual or two for the purposes of assigning branch probablities? What if we just exchange a few words? What if we just gesture at each other? What if we communicate using smoke signals? Perhaps reasoning along these lines could somehow rule out the odder "probability ghost" scenarios in a manner similar to Galileo's thought experiment about falling objects.
Huh? I don't see the linked article making value judgements.
The question as given is: suppose you go through with the above.
Do you expect to win the lottery or no?
"On the other hand as human beings we feel a duty to continue our journey. We feel a pressure to continue nation states,human species, intelligence etc."
Speak for yourself. These are parasitic memes that should never have existed and that have no valuable function for us cooperative individuals. You can count self-sacrifice for gods, nature, future generations or life in general among them also. Whenever someone calls for that, they are trying to parasitize you. The only exception on the list is probably the nation state which can be seen partly as a pragmatic cooperative of (many) individuals, e.g. to organize mutually beneficial military defense against other (aggressive) nation states. But that is highly conditional; nation states also do many things that don't benefit their citizens. Plus it needs to come with time discounting, because all individuals die in time.
Generally speaking, "moral" memes that call for self-sacrifice of individuals for the benefit of abstract systems or non-reciprocators should be thrown into the dustbin of social acceptability.
This post isn't about moral or value weight of experience, just about what is you and what you expect to see in future.
That's the first horn of the trilemma, in the link:
"Buy a ticket. Suspend your computer program just before the lottery drawing - which should of course be a quantum lottery, so that every ticket wins somewhere. Program your computational environment to, if you win, make a trillion copies of yourself, and wake them up for ten seconds, long enough to experience winning the lottery. Then suspend the programs, merge them again, and start the result. If you don't win the lottery, then just wake up automatically."
"You could say, "I have no problem with the idea that no one else could see you exerting your anthropic psychic powers, and I have no problem with the idea that different people can send different portions of their subjective futures into different realities."
I find myself somewhat reluctant to bite that bullet, personally."
Thanks for pointing out the typo; fixed now. In my model, large worlds see almost no distortion, small worlds are totally mangled, and the transition region is a small % of the total log range. Yes, tis frustrating.
Last sentence: "progeny prog" -> "progeny prob"
In your growth-drift-diffusion-absorption model which reproduces the simple Born rule inexactly, what is the explanation for our finding ourselves in a world in which there is no apparent historical discrepancy from exact compliance to that rule?
Aside: It must be frustrating to have made specific predictions which may be confirmed or refuted by theoretical analysis, and to not be able to get sufficient expert attention to move that analysis forward.
Why isn't the obvious answer that you continue on as ALL of your descendants?
I don't understand your question. It isn't crazy to be concerned about large upcoming changes, but there's a wide range between "terrified and revolt" and "feel relieved". We'll most likely muddle through in the middle.
Hello Hanson, I would like to ask you a question
As human beings we know that suffering exists. Many philosophersand philosophies have been building their ideas that it would have better never have been.
On the other hand as human beings we feel a duty to continue our journey. We feel a pressure to continue nation states,human species, intelligence etc. As an analogy we could say that we are carrying a torchlight that has been bestowed to us from previous generations and we should bestow it to next generations. It is a torchlight of the most intelligent species known.
What do you think how people would feel if we had an inception of strong AIs, EMs or transhumanistic life forms?
sqenario one:-In buddhism one basic teaching is impermanence. Even if suffering exists, impermanence gives us hope. In this way we could feel relieved. The weight of the consciousness would be of from our chest. Our suffering, status competion etc. would be impermanent and we could throw the torchlight of intelligence to some transhumanistic life forms. Now it would be transhumanistic life forms duty to continue the journey and we would have done our share.
sqenario two:-people would feel terrified and revolt
Which one of these two sqenarios is closer to your thinking?