> On 4 May 2018, at 09:20, smitra <[email protected]> wrote: > > On 03-05-2018 03:22, Brent Meeker wrote: >> On 5/2/2018 6:02 PM, smitra wrote: >>> On 02-05-2018 03:21, Brent Meeker wrote: >>>> On 5/1/2018 4:43 PM, smitra wrote: >>>>> On 01-05-2018 20:47, Brent Meeker wrote: >>>>>> On 5/1/2018 9:01 AM, Bruno Marchal wrote: >>>>>>> On 29 Apr 2018, at 19:59, Brent Meeker <[email protected]> wrote: >>>>>>> On 4/29/2018 8:53 AM, Bruno Marchal wrote: >>>>>>> But that's my question: Why isn't it the same? And even if it's not >>>>>>> how would be know? The "conscious" quantum computer assures us that >>>>>>> it not only detected that there was a welcher weg photon but that >>>>>>> it's weg was known to the "consciousness" of the quantum computer, >>>>>>> before it was erased. But why would we believe it? We already have >>>>>>> these experiments in which we know the weg was available and could >>>>>>> have been recorded, but was erased. So what is the "consciousness" >>>>>>> that adds a secret-sauce to the experiment? >>>>>>> Good question. I doubt that you can fool quantum mechanics by >>>>>>> calling it "consciousness". I think in this case the interaction >>>>>>> with the welcher weg photon would amount to sufficient decoherence >>>>>>> -- basically information was extracted that was not restored. Also, >>>>>>> of course, if the QC "forgets" what it did, how can it report on the >>>>>>> fact that it did anything. How can we believe that it actually knew >>>>>>> which slit at some point? >>>>>> Because in Deutsch experiment, not everything has been erased, notably >>>>>> the memory that he has known the result. He would say something like: >>>>>> I remember doing the measurement and writing it in the enveloppe. Now >>>>>> the envelop has been erased, and I can’t remember its content, but I >>>>>> definitely remember having known the content. >>>>>> But two questions remain. First, the empirical question of whether >>>>>> this erasure is enough to restore interference. >>>>>> I do not see why it would not been enough … in theory. You need only >>>>>> a computer able to forget a memory, but not some meta-memory that it >>>>>> has recorded a definite result. It isa bit like remembering we have >>>>>> done a dream, without being able to remember any of its content. >>>>>> In practice, that might be very difficult, if not impossible. I am not >>>>>> sure. >>>>>>> Second, why should we believe the quantum computer. >>>>>> In Deutsch proposal, it is a human. >>>>>> No, it's a conscious quantum computer. It it were a human or other >>>>>> (quasi-) classical instrument decoherence would happen when there was >>>>>> a detection of welcher weg and erasure would be impossible. >>>>>> Brent >>>>> Yes, but note that you can make that quantum computing simulation of the >>>>> observer in that thought experiment as precise as you like. You can in >>>>> principle include a simulation of the entire Earth >>>> And the outgoing EM and neutrino waves and their interaction with >>>> interstellar atoms. I'm suspicious of these fantasy thought >>>> experiments. But however detailed it may be doesn't answer my >>>> question as to what it would mean to erase the welcher weg but not the >>>> memory that the weg was detected. I noted that this is not like a >>>> classical erasure of a memory because in this case the coherence is >>>> maintained, so when the welcher weg is erased there is no long any >>>> fact-of-the-matter as to which way it went. There is no >>>> fact-of-the-matter that it was detected to go left or right. So the >>>> "memory" if it exists, is a false memory. >>>>> with billions of other people and a lot of decoherence implemented by >>>>> qubits that simulate e.g. soft photons and other environmental degrees of >>>>> freedom (and all that decoherence will end up getting reversed by the way >>>>> the computation is set up ) The point is that if computation generates >>>>> consciousness, you can in principle let any given person do the >>>>> experimental verification of the existence of multiple branches by >>>>> uploading the brain to a quantum computer and letting it be subject to >>>>> such a computation. >>>> How will the person verify it? Reversing the computation will reverse >>>> the person and erase their memory. >>>> Brent >>> It's a simple two step measurement process where you (as a virtual person >>> simulated by the QC) perform a measurement that tells you that the spin >>> (represented by a qubit) has been measured without giving you the result. >>> And then you perform the next measurement where you actually measure the >>> value of the spin component. It can then be shown that there exists a >>> unitary transform that will restore the original spin state that will >>> preserve the record of the first measurement. >> But you're speaking poetically. I, as a classical being cannot >> perform such measurements. First, how can the simulated QC person >> perform a measurement that tells you that the spin has been measured >> without giving anyone the result? In what sense is this a measurement >> of the spin, not merely a measurement of some proxy that is >> independent of the spin value? Second, what is the point of the >> second measurement "where you actually measure the spin component"; >> are you saying the first measurement did not actually measure the spin >> component even though it is supposed to tell us that it was measured? >> Third, all the techniques I've heard of for quantum erasing a >> measurement and restoring the WF are like making it so it never >> happened. You seem implicitly to take this view since you're >> concerned to preserve the record of the first measurement (which >> didn't actually measure the spin value) but not the second (which >> makes no record). >> Brent > > Yes, it's a measurement of a proxy, analogous to letting someone else measure > the spin and then that person reporting to you that the spin has been > successfully measured, without disclosing the result to you. It's not > difficult to write down a QC program to see how this works in detail. You can > take a CNOT gate a very simple observer, the control qubit is the spin that > is going to be measured (using a Hadamrd transform it can be put in a > superposition of |0> and |1>), the other bit is initialized to be in the |0> > state. We then add another qubit that changes from |0> to |1> when the gate > is applied. One can then return qubits of the CNOT gate to the original state > while leaving that extra qubit in the state it was after the measurement. > > So, the record of the measurement having taking place will be kept, while the > original spin state of, say, |0> has been restored, and that can be verified > by repeatedly carrying out this process and then also measuring the spin in > the final restored state. That final measurement always yields the same > result, proving that the qubit is indeed always restored to the |0> state. > But if the measurement at the time of the superposition were to collapse the > wavefunction, eliminating one of the two branches then the original state > would not be restored.
Very clear. > > What all of this proves, is that an observer implemented by a quantum > computer can experimentally falsify the Copenhagen Interpretation. Yes. I think that the argument showing this impossible would show that quantum computing is impossible. Bruno > > Saibal > > -- > 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 post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/everything-list. > For more options, visit https://groups.google.com/d/optout. -- 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 post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
