P != PSPACE ? and quantum computing
From: Dalibor Hrg (dalix_at_fly.srk.fer.hr)
Date: 04/19/04
- Next message: Michael Mendelsohn: "Re: Clarifying nondeterminism"
- Previous message: Mitch Harris: "Re: Clarifying nondeterminism"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Mon, 19 Apr 2004 22:52:13 +0200
Hi!
Well, I have a question about P/PSPACE problem and Deutsch-Jozsa algorithm.
Deutsch-Jozsa algorithm is a first and most simple algorithm among all
quantum algorithms. It is a problem of finding the global property of some
Boolean function of N variables. The problem requires on a classical
computer O(2^N) steps and on a quantum computer only O(N)steps. The problem
is not in P, it is in BQP which is subset of PSPACE. In this O(N) steps, the
oracle is applied only once on N qubits. Since, it is the Oracle kind of
algorithm, in literature it is written that because of this, the algorithm
can not be the proof for P!=PSPACE.
Meaning, we do not know the complexity of calculating the function in some
argument in the oracle?
What if we take that the function is really simple and requires poly number
of quantum circuits! With this assumption, we have P!=PSPACE? On classical
computer the problem with the same assumption is still not in P! On the
other side, for all other functions, we can't say if P!=PSPACE! I must say
that I'm little confused with oracles and their impact on some theoretical
questions. How important are oracles and which classical algorithms use them
for some proofs or something?
Best, D.Hrg.
- Next message: Michael Mendelsohn: "Re: Clarifying nondeterminism"
- Previous message: Mitch Harris: "Re: Clarifying nondeterminism"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|