Subquantum Mechanics & Information (contd.)
Recorded at Philosophy of Physics Discussion Group, Queens College Oxford (2002), featuring Anthony Valentini. From the Michael Wright Collection, held by the Archive Trust for Research in Mathematical Sciences & Philosophy.
- Identifier
mw0001633-cc-b_e_p- Format
- Audio recording
- Collection
- Michael Wright Collection
- Repository
- Archive Trust for Research in Mathematical Sciences & Philosophy
- Rights
- Made available for personal scholarly use. Rights in recordings are generally held by the speakers or their estates. If you believe this recording infringes your rights, please contact [email protected].
Read the automatically generated transcript
This transcript was generated by speech-recognition software from an archival recording and has not been hand-corrected. It will contain recognition errors — particularly for proper names and technical terminology — so please verify against the audio before quoting. Timestamps play the recording from that moment.
0:00 Let me, let us say this computation, I run this computer for a week, right, and at the end of the day there are 100 different eigenvalues in there that are unknown, okay, and where classically I would have to have 100 different computers running for a week, or one computer running for 100 weeks to get all these, right. Now, by doing these measurements, one can extract all the 100 answers, by having done the thing only once, for one week. Now, I'm saying, and we're assuming that these computations are being done, have an information content, if you like, that is much bigger than the work required to solve these simultaneous equations. the ratio goes as Q upon P that's my point independent of N so let us say that P is I don't know, maybe proportional to the time over which this thing is wrong whereas Q is just Q is just some things sure, it's smaller but the point is, what you're getting is just that ratio of Q and P. I mean, that's how much better you're doing. It's independent of N, the number of components in superposition. So, in a way, what is, you know, efficient about what you're doing, as I suppose, is just that instead of having to do a computation that takes P amount of time, you're doing a measurement that takes Q amount of time. And that would seem to apply equally in the classical case. process that took p amount of time and then there's a way of extracting the answer in q amount of time then sure that's a quick that's a shortcut available classically too I mean let us say let us say the output of this is the answer to some extremely complicated number theoretical problem for instance and I'm getting the answer let us say I'm getting a hundred weeks' worth of computation plus at the price of solving a hundred simultaneous equations.
2:30 Your benefit is Q over P. I haven't really talked about it. It's a fair point. If you trace one component of the state, if there was some clancy footwork involving interference effects among its expansion of space and basis then of course you can get increased efficiency that way but that doesn't seem to be that's not that's not the way we're thinking here we're thinking we're already looking at the basis with respect to which each the process is effectively a classical sequencing of digits now so it's not there that lies the exciting new possibilities you've got parallelism, but you're not able to exploit that parallelism without, you know, doing end times the amount of work you'd extract from one of them. I mean, it would be, I mean, I would be worth looking at, I mean, this is the sort of most simple-minded way of doing it. I use this, there's different branches, doing different computations, and I figure out all the bandages by looking at the hidden variable. Presumably, one could do much cleverer things, but it just hasn't been developed. What would happen if one looked at, say, some of the clever quantum computational schemes, and then you assume you also were able to look at the hidden variables and see that information back in? I mean, there's a whole field there that no one has gone to. So, but I mean, I agree, this is a fair comment, but this is also, it's working out to be liberal to man. Something you'd probably be able to do is in those quantum algorithms which just give you the answer with a certain, the right answer with a certain probability, and then you're you probably might well, using this sort of scheme, be able to not just run it once. But then that's not good enough. Well you might have interesting algorithms that actually give you infinitesimum small probability. But with, if there's no equilibrium matter, one could target that, that very low probability,
5:00 and that might open up a high amount. I'm sure, I'm sure, I just haven't looked at it. I have another question which is about, couldn't you say you believe sort of the general picture here of having noise, hiding what's going on from us rather than believing the details bone theory, what sort of other possible theories do you have in mind apart from that or I mean what force is that stepping back in you if the reason is for not believing in the bone why would you not believe in any others either? I don't have any particular, I mean the deploy bone theory seems nice and simple and seems to me that in the state of ignorance that we have, one may as well work with that, but if someone comes along with a different theorem, I'm happy to look at it. But what I believe more is the general scenario, that there is some hidden variables theorem, and we are looking at a particular equilibrium state of that theorem. And I mean, as I've shown, there is a general theorem now on any deterministic theory out of equilibrium you get signaling to suggest to me that we're in this finely tuned state and any deviation from that state unleashes this stuff that is hidden from I suspect that there are, I mean now as I said before, now there are three properties of the Broybone theory that are properties of any hidden variable theory, non-locality, contextuality, and this signal locality theory. One suspects that there may be other things in Broybone theory that are actually general functions. Perhaps something like this, for instance, to factor that. Is it true in any deterministic hidden variable theory? Out of
7:30 narrow, non-equilibrium distribution centered on particular hidden variables. If you could access that, you could read all the results of a parallel computation. I don't know. So, I don't know how much of the, I mean, when you look at the de Broglie-Bohm picture and figure out these things that one could do, how much of it is true in general and how much the feature of that theory is hard to tell, but there's certainly a core that is genuine, but there's an underlying non-locality, hidden in equilibrium, other equilibrium you get signaling, um, you know, there's nothing like, no one has proved an H-theorem or relaxation or anything for an arbitrary deterministic invariance. How do those theories, the thermoistic hidden variables theories that also satisfy the three, concerning what sort of things those are like, they're not referring to, is that true? Um, well, um, to be honest, I mean, I'm not saying, Um, a general hidden, I mean a hidden variable theory that say reproduces general non-autonomistic quantum mechanics, it's deterministic, but is not the Breuer-Bohm theory. Um, is it strictly equivalent between a certain limit that reproduces, I mean Lee Smolin now has some new non-autonomical hidden variable theory? It is deterministic. Is this matrix physical as matrix elements which are hidden variables in some sense? Yes, the diagonal elements are the hidden variables that possess values. The action principle is respect to an external classical time. Otherwise it's geometry-free. It just involves a cometary of these matrices. And if you look at the effective action on the diagonal elements, due to the off-diagonal elements, you get out equations which a complicated limiting process could be coding at. Sure, now whether in that limiting process one has approximations that you can find some regime in which they're not satisfied, I'm not sure.
10:00 And whether it has these other properties, The non-locality is pretty explicit, the signaling is not clear, and the textuality is not clear. None of those things are not clear. And look, there's so many steps in the argument, I mean, I think they need to be worried that they think this is part. The more trivial quantification of the Brueggemann theory is to add another term to the velocity whose divergence, well, if only one multiply by the psi-squared, the divergence is more advantageous and it would still satisfy the same continuity. the way I think of it personally is it would probably well probably let's say with a bit of luck maybe we're in the situation people were in in the late 19th century where they're thinking about gases and you've got more picture molecules little billiard balls banging around inside a container simple theories that have had some truth to them but i hope the pilot web theory has some basic features which are true at least qualitatively um but i mean no wonder if i how can i possibly know if we can't see we're stuck in this spectacular you can't see of course no one can know the contrary you have to find this non-equilibrium really to make programs Thank you.
Transcript not yet available for this recording.