Classical and Quantum Computers

Brian J Flanagan wrote
>  Biological systems, from a physical standpoint, just are
large quantum fields; ergo all their processes are inherently
quantum, including those processes regarded as
computational.

Matthew Donald wrote:
This is debatable, except in the trivial sense in which quantum field
theory can used as the basis of description of all physical systems.

bj:
If consciousness is a kind of field process, then that would seem
nontrivial, yes?

md:
  In my work on quantum theory and consciousness, I have always started by
remarking that brains are warm and wet.  By this I mean to imply that I
do not believe that there is anything exceptional about the physical
interactions going on inside our heads.  Brain states are quantum
states, but they are thermal states; they are like the states of the
warm coffee in the mug on my table rather than like the states of a
superconductor.

bj:
Brain states are highly ordered, unlike coffee. Transistors are also
q-computational devices, and do not require low temperatures.

md:
Umezawa, as mentioned by Flanagan,  took a different point of view.
Umezawa thought, for reasons I have never understood, that the
conventional neuron doctrine was insufficient to explain memory.

bj:
He was quite taken with the kind of nonlocality of memory noted by Pribram:

"For many years it was believed that brain function is controlled solely
by the classical neuron system which provides the pathway for neural
impulses. This is frequently called the neuron doctrine. The most
essential one among many facts is the nonlocality of memory function
discovered by Pribram ..."

Umezawa, 'Advanced Field Theory'


There is nothing at the neural level of comparable simplicity to the
primary & secondary qualities disclosed in observation, whereas physics is
all about primary qualities. Then, too, the operator formalism is
remarkably amenable to the secondary qualities, as is most readily seen in
the case of color and sound.


Flanagan wrote:
> quantum computation research is very new, and may well
> continue to reveal all sorts of remarkable features of
the quantum realm.  That realm is only incompletely
understood

mb:
Indeed.

But there has been a significant change in quantum computation research
over the last decade.  What I have been trying to do in this discussion
has been to try to get the nature of that change across.  The essence of
it is that a standard idea of a quantum computer has come to be accepted.

bj:
Copenhagen was also accepted, and for a number of decades. The standard
idea of q-computation is a nice start, but I would venture to say it only
scratches the surface.

md:
The field may be less open, perhaps less exciting, than it once was, but
it is much more serious and very productive (at least in terms of papers).

bj:
I am reminded of JS Bell's remark that what no "hidden variables" proofs
demonstrate is a lack of imagination. And then, Bohr did not write much,
and what he did was famously obscure -- even, in one notable instance, to
him -- but clearly his work was both serious and productive. I don't wish
to seem contentious, or to launch into a polemic, but the careerism
evident in the "publish or perish" syndrome often seems neither serious
nor productive, and results in vast quantities of stuff being churned out,
much of it of questionable value.

md:
But let's go back even further, and suppose that we had
just got to the telegraph relay.  [...]  That physics,
in terms of metastable states for example, is enough to let
us see that neurons can function like mechanical switches.

bj:
Leaving us at square one, in Leibniz' mill: How does that model account
for perceptions? The state space is far too small to accommodate the
variety of sensory qualities.

md:
A qubit is simply a pure quantum state in a two-dimensional
Hilbert space.   Work with relays showed us that mechanical
switches were fairly easy to construct and to control.  Work with
elementary quantum computers is showing us that it is very difficult to
keep the quantum states of systems of any complexity pure; let alone to
control them in the ways that would be required for any sort of advance
beyond classical computation. I would actually go so far as to argue that
there are no systems involved in processing information anywhere in the
brain which ever maintain pure quantum states; let alone any which have
the mechanisms required to control those states.

bj:
In the case of the quantum brain, it is wholly unclear to me what pure
states and/or long range coherence might achieve that simple superposition
cannot, aside from the possibility of very fast, nonlocal computation.
I.e., the photon fields must couple with the electron fields to achieve
causal efficacy with respect to the larger brain & organism.

Flanagan wrote:
> On the field level, the exchange of photons in the IR
region is not essentially different from photon exchanges at
other frequencies.

md:
For any frequency range you fancy, you can write down a
Hilbert space to model the states of photons in that range and you
can model the electromagnetic interactions which affect those
states.

bj:
Notice that the secondary qualities also lend themselves to a vector
representation, and that their observed values depend on QM phase
relations.

md:
The state of the electromagnetic field produced by the motions of a
million or more ions reflects those motions, and is just as impure and
thermal and uncontrollable as the state of the ions themselves.

bj:
Thermal states are not controllable? Wait till the utility companies get
wind of this. Then, too, we regularly control ions with EM fields. Then
again, just how thermal matter is within the confines of a cell seems open
to question. Finally, if by thermal you mean random, I would argue that
this is only a statistical measure of processes which may well be
determinate in character, right down to the quantum level.