On Thu, Aug 1, 2019 at 5:57 PM Bruce Kellett <[email protected]> wrote:
> On Fri, Aug 2, 2019 at 2:32 AM Jason Resch <[email protected]> wrote: > >> On Mon, Jul 29, 2019 at 6:08 PM Bruce Kellett <[email protected]> >> wrote: >> >>> On Tue, Jul 30, 2019 at 6:32 AM Jason Resch <[email protected]> >>> wrote: >>> >>>> On Mon, Jul 29, 2019 at 6:41 AM Bruce Kellett <[email protected]> >>>> wrote: >>>> >>>>> On Mon, Jul 29, 2019 at 8:43 PM Jason Resch <[email protected]> >>>>> wrote: >>>>> >>>>>> On Fri, Jul 26, 2019 at 6:29 PM Bruce Kellett <[email protected]> >>>>>> wrote: >>>>>> >>>>>>> On Sat, Jul 27, 2019 at 8:44 AM Jason Resch <[email protected]> >>>>>>> wrote: >>>>>>> >>>>>>>> On Fri, Jul 26, 2019 at 2:24 AM Bruce Kellett < >>>>>>>> [email protected]> wrote: >>>>>>>> >>>>>>>>> >>>>>>>>> MWI is irrelevant to this discussion, since the branches in MWI >>>>>>>>> are completely disjoint and form separate coherent worlds. Without >>>>>>>>> overlap, >>>>>>>>> common sense notions of personal identity continue unchanged in all >>>>>>>>> branches separately. >>>>>>>>> >>>>>>>> >>>>>>>> Can we really ignore the global view? >>>>>>>> >>>>>>> >>>>>>> Yes, we have no evidence that such a view exits or even makes any >>>>>>> sense. >>>>>>> >>>>>> >>>>>> Why doesn't it? >>>>>> >>>>> >>>>> Because there is no "view from outside" as it were. >>>>> >>>> >>>> But this is exactly the situation for Wigner's friend, or when we use a >>>> quantum computer. >>>> >>> >>> The trouble is that Wigner's friend, or the cat for that matter, is a >>> classical, decohered, concept. So there are no quantum superpositions >>> involved. The fact that you can write the situation for Wigner's friend in >>> the form of a superposition does not mean that it actually is one. One >>> cannot interfere the different components. >>> >> >> From the standpoint of Wigner, or the cat in the box, it remains in a >> superposition so long as it is isolated from the environment, does it not? >> > > No, because both Wigner's friend and the cat are decohered objects -- not > in superpositions because the two states can no longer interfere . > They have decohered locally (within the steel box), but not with their surrounding environment. This is the essence of how quantum computers work, the qubits can interact locally while performing the computation, because the whole system is isolated from the rest of the environment. > > What theory or interpretation of QM are you assuming when you say no >> superposition are involved? >> > > The theory of quantum mechanics that follows through to give the results > of experiments without stopping half way. > But you think it stops somewhere, since you don't think it can be extrapolated to observers or the universe as a whole. > > > It would also be the case for any being in another universe choosing to >>>> analyze ours using the power of simulation. Do you not believe in an >>>> objective reality? >>>> >>> >>> Objective reality is that obtained by intersubjective agreement. By this >>> token, the "global wave function" has no objective existence. >>> >> >> What if Everett and Feynmann agree there is a global wave function, and >> choose to study the implications of that theory? >> > > Argument from authority. Both Everett and Feynman were frequently wrong > about all sorts of things. > I think you miss my point. There was "inter-subjective agreement" between those two concerning the reality of the global view. Under your requirements, it then meets your definition of objective. > > > We speak of a superposition of a wave function whenever we speak of a >>>>>> system separated from an external environment. Can we not view the whole >>>>>> universe this way? Feynman and Everett thought we could, and Wheeler >>>>>> thought so on Tuesdays. >>>>>> >>>>> >>>>> Argument from authority is no argument in this case, because it makes >>>>> no sense to view the whole universe in this way. There is no "person" who >>>>> sees the global wave function. >>>>> >>>> >>>> Nor is there any person who sees beyond the horizon of a black hole. >>>> But we can use our best theories to describe that is happening there. Why >>>> can't we do the same for the global view? >>>> >>> >>> Poor analogy. We can readily see what is beyond the horizon of a black >>> hole -- just jump in and look. We can't do anything to obtain your required >>> "global view" because we are restricted to having only "relative state" >>> views, even with Everett. >>> >> >> If you acknowledge they are relative, then what are they relative to? >> The other views. Is it not worth ascribing a name to the total set of >> mutually relative views? >> > > The state is relative to a particular observer. There is no point is > thinking about the total set of mutually orthogonal relative views because > they can have no influence on each other -- they are outside the realm of > experimental physics. > That doesn't mean there is no point thinking about them. Cosmologists anywhere will talk about the universe beyond the cosmological horizon, despite it having no impact on any experiments. They talk about it because it is a necessary consequence of leading and testable theories. > > > According to Everettt, persons and quantum phenomena are relative to >>>>> particular branches of the wave function. >>>>> >>>> >>>> Is there a wave function for the universe as a whole? >>>> >>> >>> Who knows. I find the concept somewhat dubious -- over-enthusiastic >>> extrapolation from simple laboratory experiments. >>> >> >> So you think Schrodinger's equation is wrong? >> > > Why do you think that I might think this? > You think it is "overly enthusiastic" to apply it to the universe as a whole. This implies you suspect it stops working at a certain point. > > >> Often the most marvelous consequences revealed themselves through simple >> laboratory experiments. >> > > The Schroedinger equation has only ever been tested in simple quantum > systems -- we have no evidence that it could apply to the whole universe. > We could not even begin to write such a thing down in any useful way. > We have no evidence that it can't be applied to the whole universe. What is the basis for your suspicion that it cannot be applied in this way? > > > What if we want to consider cases like Wigner's friend? Or cases where we >>>>>>>> emulate brains in quantum computers? >>>>>>>> >>>>>>> >>>>>>> What about such cases? Despite David Deutsch, these do not prove the >>>>>>> truth of MWI. >>>>>>> >>>>>> >>>>>> No but it casts significant doubt on the single world view, which >>>>>> cannot account for them. >>>>>> >>>>> >>>>> The single world view can account for these easily, so such cases >>>>> provide no evidence for MWI. >>>>> >>>> >>>> Where do the 10^1000 intermediate computational states occur in the >>>> single universe view of a quantum computer factoring a big semiprime >>>> number? >>>> >>> >>> Quantum computers work by interference of quits, and such interference >>> can only take place in one world -- different worlds are orthogonal. The >>> fact that one can analyse a quantum computer in a particular basis which >>> can be represented as a series of parallel computations does not mean that >>> this is actually what happens. Heuristic constructs seldom correspond to >>> reality. >>> >> >> None of this comes anywhere close to addressing my question. >> > > Well, you have either not understood the question, or my answer to it. > I asked where those 10^1000 intermediate computation states are realized, and your reply was a basic description of how quantum computers use qubits and interference. You said this all takes place in one world, but the total information content and computational capacity of the observable universe about 800 orders of magnitude less than 10^1000. You then added a sentence that suggested the intermediate computational states perhaps don't exist, but then how does the correct answer get into the output bits when we read it? David Deutsch said he has never seen a sensible answer to the question of how quantum computers work from the context of any single-universe interpretation. Do you think your answer would satisfy him? > > > Superpositions exist in a single world. If we form multiple worlds via >>>>>>> decoherence, then the worlds are, by definition, orthogonal, so there >>>>>>> is no >>>>>>> possibility of their ever recombining. David Deutsch got this wrong many >>>>>>> times. Quantum computers work by interference of qbits -- so they must >>>>>>> all >>>>>>> exist in the same world. A conscious quantum computer does not >>>>>>> experience >>>>>>> other worlds. >>>>>>> >>>>>> >>>>>> If you fed in as input, qubits prepared to be in a superposition >>>>>> using Hadamard gate, then the conscious mind would experience many >>>>>> states, >>>>>> simultaneously. >>>>>> >>>>> >>>>> Demonstrate that by actually doing it. >>>>> >>>> >>>> I described the process. We lack the technology to do this today, but >>>> it is exactly what our current theories suggest would happen. >>>> >>> >>> Suggestions are not reality, and I doubt that the theories do actually >>> suggest this. >>> >>> >>>> Scott Aaronson, an expert in quantum computing agreed that if the >>>> computer program is conscious, then one so-prepared on a quantum computer >>>> would realize multiple conscious states simultaneously. >>>> >>> >>> Argument from authority. Scott Aronson can provide no more evidence for >>> this that you can. >>> >> >> I am not saying it is true because Scott Aarson says it is true, I am >> saying it is a consequence of the theory (and if there were an obvious flaw >> in the reasoning it is likely Scott would have identified it). >> > > It is a consequence of an unevidenced extrapolation of the theory to a > realm in which it has not been tested. Past experience in physics shows > that this is not a reliable approach. > Quantum computers keep growing in capacity. I think Google is now up to a 200 qubit quantum computer. Aaronson has bet $100,000 of his own money on that large scale quantum computers are possible. If you can show him why he is wrong, that money is yours. Do you have a guess for the number of qubits where no further progress is possible? > > >> You have not indicated any flaw in the reasoning; only expressed your own >> personal doubt. On what those doubts are based, I have no idea. >> >> >>> >>> >>>> Can you show any flaw in the reasoning? >>>> >>> >>> Consciousness is a classical phenomenon and as such cannot participate >>> in a true quantum superposition. >>> >> >> If consciousness is the result of a classical computation (as I assume in >> the above reasoning), then the superposition realizes those multiple >> parallel classical computations, hence you get the multiple points of view. >> > > You do not have multiple parallel classical computations in a quantum > computer. That is not how the algorithms work. > It depends on the algorithm. If you want all or most branches to end up with an approximate answer, you need to design the algorithm such that incorrect answers destructively interfere. But the most straightforward quantum computation, which is merely a classical algorithm executed on qubits that are in a randomized superposition, can be viewed as many parallel computations. Of course you will end up with the output of a function with a random input, which may or may not be useful, but before you read the output it is in a superposition of the "range" of that function. > > >> The fact that anything done by a quantum computer can also be done by a >>> classical Turing machine proves this beyond any doubt: you cannot have >>> superpositions in a classical machine. You can simulate them, but then a >>> simulation of a rain storm does not make you wet. >>> >> >> The superposition is irrelevant for consciousness, but the superposition >> is relevant to the thought experiment. The superposition is what implies >> the multiple unique states of achieved by the computation, and these >> multiple states create multiple unique conscious viewpoints. >> > > The assumption of a superposition is wrong. I have given many reasons why > this is so. You cannot get a quantum superposition of decohered classical > objects. > Decoherence is just the spreading of a superposition to involve more particles. > Consciousness is necessarily a decohered classical object since it > involves memory. Memory is irreversible, so cannot arise by purely unitary > evolution. > Quantum computers use reversible logic gates so that bits can be stored and processed without shedding heat into the local environment and destroying the superposition before the computation is complete. > If the person that comes out of a quantum computer can have no memory of > being in a superposition, why do you worry about it? There is in principle > no way you could ever find out about it. > The experiences are still created and had, even if the persons are destroyed in the end. The fact that WW1 soldiers have all died doesn't mean they never suffered in the war. > > To make my example more clear, let's assume a simple neural net trained >> for image recognition manifests some basic form of "protoconsciousness". >> We then initialize a set of 10,000 input qubits to each be in a >> superposition of (1 and 0). These input registers represent a 100x100 pixel >> black and white image. The neural network is then run on the quantum >> computer and processes the random image for image recognition, and matches >> it to the word that scores most highly from what it sees and places this >> word in an output register to be read. >> >> Do you agree that: >> >> 1. There are 2^10,000 possible results from reading the 10,000 qubits? >> 2. The execution of the neural net on the quantum computer processes each >> of these 2^10,000 possible images? >> 3. Before reading the output register, those output bits are in a >> superposition of every possible word the neural net is trained to identify? >> 4. If "protoconsciousness" can be ascribed to this neural net, then this >> consciousness perceived each of the 2^10,000 possible images? >> > > I do not agree with any of the above. Consciousness cannot enter into any > verifiable superposition. > > So you disagree with the statements numbered "1", "2", "3" as well? Jason -- 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/CA%2BBCJUjZ%2BmQDc%3DyGOJ9sP0F%3DBsB-nnUWqfSUYtFit9KKQNVrcg%40mail.gmail.com.
