On Wednesday, July 7, 2021 at 9:29:24 PM UTC-5 Brent wrote:
> > On 7/7/2021 2:04 PM, Jason Resch wrote: > > > > On Wed, Jul 7, 2021 at 3:43 PM 'Brent Meeker' via Everything List < > [email protected]> wrote: > >> >> On 7/7/2021 10:09 AM, Jason Resch wrote: >> >> >> >> On Wed, Jul 7, 2021 at 11:53 AM 'Brent Meeker' via Everything List < >> [email protected]> wrote: >> >>> >>> On 7/7/2021 2:24 AM, Jason Resch wrote: >>> >>> >>> >>> On Wed, Jul 7, 2021, 12:14 AM 'Brent Meeker' via Everything List < >>> [email protected]> wrote: >>> >>>> >>>> On 7/6/2021 6:50 PM, Jason Resch wrote: >>>> >>>> >>>> >>>> On Tue, Jul 6, 2021, 9:39 PM Bruce Kellett <[email protected]> wrote: >>>> >>>>> On Wed, Jul 7, 2021 at 11:29 AM Jason Resch <[email protected]> >>>>> wrote: >>>>> >>>>>> On Tue, Jul 6, 2021, 4:07 PM 'Brent Meeker' via Everything List < >>>>>> [email protected]> wrote: >>>>>> >>>>>>> On 7/6/2021 10:34 AM, Jason Resch wrote: >>>>>>> >>>>>>> On Tue, Jul 6, 2021 at 12:27 PM 'Brent Meeker' via Everything List < >>>>>>> [email protected]> wrote: >>>>>>> >>>>>>>> And you're never going to find a being that behaves intelligently >>>>>>>> based on information that can be quantum erased. >>>>>>>> >>>>>>> You need only a quantum computer with enough qubits. >>>>>>> >>>>>>> Can you prove that? How does this quantum intelligence ever arrive >>>>>>> at a definite decision? >>>>>>> >>>>>> >>>>>> Prove? No. But I think I can justify it: >>>>>> >>>>>> 1. Quantum computers are Turing equivalent, they can compute anything >>>>>> a classical computer can. >>>>>> >>>>>> 2. Human brains are believed to operate according to physical laws, >>>>>> all known of which are computable. >>>>>> >>>>>> 3. Humans are conscious. >>>>>> >>>>>> 4. By any of: Chalmers's principle of "Organizational invariance", or >>>>>> "multiple realizability", or the "Generalized Anti-Zombie Principle", or >>>>>> the "computational theory of mind", a functionally equivalent >>>>>> computation >>>>>> to that of a conscious human brain will be equivalently conscious to >>>>>> that >>>>>> brain. >>>>>> >>>>>> 5. Quantum computers are reversible. >>>>>> >>>>>> By 1 & 2, a quantum computer can simulate a human brain. By 3 & 4, >>>>>> such an emulation will be conscious. By 5 any computation performed by a >>>>>> quantum computer can be quantum erased by reversing the circuit back to >>>>>> its >>>>>> starting state. >>>>>> >>>>>> It reaches a definite decision by virtue of completing its processing >>>>>> before ultimately being reversed. This prevents an outside observer from >>>>>> learning the decision, but it's made nonetheless during the course of >>>>>> the >>>>>> processing. >>>>>> >>>>> >>>>> How do you know that it has reached a definite decision? Without >>>>> having it print out some irreversible record? If it prints out a >>>>> (pseudo-)classical record, the initial state is not recoverable. >>>>> >>>>> Bruce >>>>> >>>> >>>> By either: >>>> >>>> 1. Analyzing the circuit >>>> >>>> But the question is whether such a circuit is possible. >>>> >>> Do you disagree with any of the five premises I defined above? If not do >>> you see a flaw in my reasoning or conclusions? If not, then why shouldn't >>> such a circuit be possible? >>> >>> This what I find dubious: *"It reaches a definite decision by virtue of >>> completing its processing before ultimately being reversed. This prevents >>> an outside observer from learning the decision, but it's made nonetheless >>> during the course of the processing." * First, I doubt that it both >>> reach a definite decision and have that quantum erasable. >>> >> If you doubt it reaches a certain definite decision state, you could >> interrupt the quantum computer midway through its processing and entangle >> yourself with one of its superposed states to verify that the AI/mind was >> in a state of having reached a definition conclusion. >> >> >> ?? If I do that by entangling with a superposition, then I either >> collapse it or "I'm of two minds". >> > > Yeah you spoil the process by interrupting it early, but it lets you > verify the computation reaches those intermediate states in the course of > its normal evolution, including in those that you allow the algorithm to > run to completion. > > >> >> >> >>> Second, you've made "decision" something internal. Intelligence >>> requires acting in the world. >>> >>> >>> The environment for this AI are the qubits initialized as the input to >> the mind. It acts in this world by performing actions that ultimately >> affect the output of this quantum computation. >> >> >> My original point was, "And you're never going to find a being that >> behaves intelligently based on information that can be quantum erased." In >> the environment A=0, B=0, and any other set of A, B values the algorithm >> outputs B=1 and then erases it. Is this intelligent behavior? >> > It's perhaps a thermostat level of intelligence, but you can make it > arbitrarily complex, as in Deutsch's AI example that does the same thing as > this simple circuit. > > No matter how complex you make it (and maybe because you make it complex) > you cannot both act on it and quantum erase it. There's a reason that > intelligent beings live in a quasi-classical world. They would never > evolve in a world that was reversible. And as Bruce points out, this world > is not just statistically irreversible, it's inherently irreversible > because all but a finite part is receding faster than the speed of light. > > Brent > The accelerated expansion of the universe does not contribute to entropy. The role of quantum decoherence is phenomenologically evident, in that with wave function collapse or reduction quantum coherent states become a statistical decoherent set. In one way or another quantum entanglement is lost and qubits are either local due to MWI splitting, maybe there is real collapse as in GRW or some other thing. I have a different way of looking at this that does not involve these interpretations. LC -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/b00b8baf-c2c2-4a81-a168-07259ecbd0den%40googlegroups.com.
