The Quantum Setting with Randomized Queries for Continuous Problems
The standard setting of quantum computation for continuous problems uses deterministic queries and the only source of randomness for quantum algorithms is through measurement. Without loss of generality we may consider quantum algorithms which use only one measurement. This setting is related to the worst case setting on a classical computer in the sense that the number of qubits needed to solve a continuous problem must be at least equal to the logarithm of the worst case information complexity of this problem. Since the number of qubits must be finite, we cannot solve continuous problems on a quantum computer with infinite worst case information complexity. This can even happen for continuous problems with small randomized complexity on a classical computer. A simple example is integration of bounded continuous functions. To overcome this bad property that limits the power of quantum computation for continuous problems, we study the quantum setting in which randomized queries are allowed. This type of query is used in Shor’s algorithm. The quantum setting with randomized queries is related to the randomized classical setting in the sense that the number of qubits needed to solve a continuous problem must be at least equal to the logarithm of the randomized information complexity of this problem. Hence, there is also a limit to the power of the quantum setting with randomized queries since we cannot solve continuous problems with infinite randomized information complexity. An example is approximation of bounded continuous functions. We study the quantum setting with randomized queries for a number of problems in terms of the query and qubit complexities defined as the minimal number of queries/qubits needed to solve the problem to within ɛ by a quantum algorithm. We prove that for path integration we have an exponential improvement for the qubit complexity over the quantum setting with deterministic queries.
KeywordsQuantum computation randomized queries qubits complexity
Unable to display preview. Download preview PDF.
- 1.Bakhvalov N.S. (1959). On approximate calculation of integrals (in Russian), Vestnik MGV. Ser. Mat. Mekh. Aston. Fiz. Khim. 4:3–18Google Scholar
- 2.Beals R., Buhrman H., Cleve R., Mosca R., and R. de Wolf, Quantum lower bounds by polynomials, Proc. FOCS’98 352–361 (1988). Also http://arXiv.org/quant-ph/9802049.Google Scholar
- 11.Heinrich S., Kwas M., and Woźniakowski H. (2004). Quantum Boolean Summation with Repetitions in the Worst-average case Setting. In: Niederreiter H. (eds) Monte Carlo and Quasi-Monte Carlo Methods 2002. Springer Verlag, Berlin, pp 243–258Google Scholar
- 12.Heinrich S., On the power of quantum algorithms for vector valued mean computation, submitted for publication, 2004. See http://arXiv.org/quant-ph/04031109.Google Scholar
- 13.Heinrich S., Novak E., and Pfeiffer H. How many Random Bits do we Need for Monte Carlo Integration?. In: Niederreiter H. (eds) Monte Carlo and Quasi-Monte Carlo Methods 2002, Proceedings of a conference held at the national University of Singapore, November, 2002 (Springer, Berlin, 27–49).Google Scholar
- 18.Nayak A., and Wu F., The quantum query complexity of approximating the median and related statistics, in Proceedings of the 31th Annual ACM Symposium on the Theory of Computing (STOC), pp. 384–393, 1999. Also http://arXiv.org/quant-ph/9804066.Google Scholar
- 19.Niederreiter H., Random number generation and Quasi-Monte Carlo methods, in Proc. vol. 63 of SIAM CBMS-NSF Regional Conference Series in Applied Mathematics (SIAM, Philadelphia, 1992).Google Scholar
- 23.Novak E., and Woźniakowski H. (2001). When are integration and discrepancy tractable?. In: DeVore R.A., Iserles A., Süli E. (eds) Foundation of Computational Mathematics, Oxford, 1999. Cambridge University Press, Cambridge, pp. 211–266Google Scholar
- 31.Traub J.F., A continuous model of computation, Phys. Today May, 39–43 (1999).Google Scholar