Quantum computers may end up being not-more-advantageous for "quantum modeling" than the current trend in using DLNs (deep leaning nets).
e.g. https://arxiv.org/abs/1909.02487 @philipthrift On Tuesday, October 6, 2020 at 4:57:45 AM UTC-5 Lawrence Crowell wrote: > On Monday, October 5, 2020 at 1:55:20 PM UTC-5 [email protected] wrote: > >> On Sun, Oct 4, 2020 'Brent Meeker' via Everything List < >> [email protected]> wrote: >> >> On 10/4/2020 12:52 PM, Lawrence Crowell wrote: >>> >>> >> Quantum computers, or processors, will make more inroads into >>> things. They have a possible big role in understanding quantum black holes >>> and quantum complexity. Any NP problem can be worked faster, at least in >>> principle, with a quantum computer. >>> >>> * > I don't think there's any proof of that. Given any quantum computer >>> algorithm, it is possible that there is an equally fast classical algorithm* >>> >> >> It's true that although a quantum algorithm has been found that can >> factor numbers efficiently there is no proof a classical algorithm cannot >> be discovered that would do the same thing, in fact it has never been >> proven that P≠NP, although nearly all mathematicians believe that is the >> case. However it has been proven that a recently discovered exotic class >> of problems can be solved In polynomial time but even if it turns out to >> everybody's surprise that P=NP and a classical algorithm is found to make >> use of that fact a classical computer could never do as well solving them >> as a quantum computer. It's so new that nobody is yet quite sure if this >> exotic class of problems is of interest in themselves or is interesting >> only because a conventional computer could not solve them efficiently but a >> quantum computer could. Although falling short of a proof it gives yet more >> ammunition to those who believe a quantum computer can solve more familiar >> practical problems faster than a classical computer ever will be able to. >> >> Oracle Separation of BQP and PH >> <https://eccc.weizmann.ac.il/report/2018/107/> >> >> I think the killer application for a quantum computer will be simulating >> quantum systems. >> >> John K Clark >> >> The physical idea is that a quantum computer is a faster is that in > principle if we had quantum brains it would really be exponentially faster. > However, the result of a quantum computation can only be be manifested if > the entanglements are decoded by a classical signal. This "undoes" the > exponential speed up. However, for teleporting a state with a Bell pair the > classical part has half the information. Then in principle, for quantum > computing there is a speed up that is some fraction of what occurs with a > classical computer. The actual speed up is dependent on the algorithm as > well. > > This paper on BQP and PH made a related point in that BQP has the need for > fewer oracle inputs, which is the same as saying user inputs. This means > for a range of problems quantum computing will have an economy of time or > scale. > > Certainly right away quantum computing will be mostly used for modelling > systems, in particular quantum systems. Quantum computing has analogues > with black holes as well. The complexity of computing has analogues with > quantum complexity of black holes. We may then have laboratory-like forms > or simulations of black holes with quantum computers. In fact I think this > can happen with a certain optical process with atoms that can make an atom > quantum computing. > > 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/62ab81f8-9804-48ca-be92-c93abfabf017n%40googlegroups.com.
