CO481/CS467/PHYS467 ASSIGNMENT 1
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
CO481/CS467/PHYS467 ASSIGNMENT 1
Due Friday March 17th at 8pm, electronically using Crowdmark
(will constitute 12% out of the 60% assignment marks)
1. Modular arithmetic and factoring 3 marks Let r be the order of 3 mod 35.
(a) Find r .
(b) What is 3601 mod 35?
(c) Find GCD(35, 3 − 1) and GCD(35, 3 + 1).
2. Quantum searching 3 marks
(a) Find the smallest positive p so that quantum search via amplitude amplification finds a solution with certainty using two iterations of the quantum search iterate? Also give a three decimal approximation to p.
(b) Suppose we have a quantum algorithm A that produces a solution to f(x) = 1 with probability . What is the smallest positive integer k so that k + 1 iterations of the quantum search iterate finds a solution with probability less than k iterations would?
3. Exact one-out-of-four searching 2 marks
Let f : {0, 1}n '→ {0, 1} . Suppose we wish to find a string x ∈ {0, 1}n such that f(x) = 1. Suppose further that exactly one quarter of all the strings x in {0, 1}n satisfy f(x) = 1.
Show how to find a string x with certainty using exactly one evaluation of the black-box Uf : |x⟩|b⟩ '→ |x⟩|b ⊕ f(x)⟩ .
4. Quantum counting 3 marks
Suppose f : {0, 1}2n '→ {0, 1} with a promise that the number of solutions to f(x) = 1 is either 2n or 2n+1 .
Explain how to decide, with high probability, whether there are 2n or 2n+1 solutions using
a number of queries to Uf : |x⟩ '→ ( − 1)f(x)|x⟩ in O(^2n ).
5. Collision-finding 4 marks
Let f : {1, 2, . . . ,N} → X for some finite set of strings X, with the property that f is two-to-one. That is, for each value y occurring in the range of f, there are two distinct inputs, x1 ,x2 such that f(x1 ) = f(x2 ) = y .
Suppose you are given a black-box for implementing Uf : |x⟩|b⟩ '→ |x⟩|b ⊕ f(x)⟩, where x ∈ {1, 2, . . . ,N} and b ∈ {0, 1} .
Consider the following collision-finding algorithm:
❼ Query f(1),f(2), . . . ,f(M), for some M << N .
❼ If f(x1 ) = f(x2 ) for distinct x1 ,x2 ∈ {1, 2, . . . ,M}, then output the collision pair (x1 ,x2 ).
❼ Otherwise, perform a quantum search for a value x2 ∈ {M + 1,M + 2, . . . ,N} such
that f(x2 ) = f(x1 ) for some x1 ∈ {1, 2, . . . ,M} . Output (x1 ,x2 ).
(a) Assuming f(1),f(2), . . . ,f(M) are distinct, what is the probability p that a value x sampled uniformly at random from {M +1,M +2, . . . ,N} will satisfy f(x) = f(x1 ) for some x1 ∈ {1, 2, . . . ,M} .
(b) How many quantum queries does this algorithm need in order to find a collision with constant probability? Express your answer in terms of N and M and using big-O notation. (Do not forget about the queries to compute f(1),f(2), . . . ,f(M) in the first step.)
(c) Let M = Ne for some constant ϵ > 0. Find the value of the constant ϵ that minimizes the number of queries (up to constant factors) needed to find a collision with high probability.
6. Parallel quantum searching 2 marks
Suppose f : {0, 1}n f→ {0, 1} has exactly one solution.
Let Uf : |x⟩ f→ ( − 1)f(x) |x⟩ .
Show how to find the unique solution to f(x) = 1 with high probability using O(√2n−k) applications of the operation Uf ⊗ Uf ⊗ · · · ⊗ Uf = U
7. Parallelizing phase-queries 3 marks
Let Uϕ denote the unitary operation that maps |0⟩ f→ |0⟩ and |1⟩ f→ eiϕ |1⟩ .
Note that Ukϕ = Uϕ(k) . However, if a black-box process for implementing Uϕ takes time t then implementing Ukϕ in this serial way takes time kt.
(a) What is the result of applying Uϕ ⊗Uϕ ⊗ · · · ⊗Uϕ = Uk to |00 ... 0⟩+ |11 ... 1⟩ = |0⟩⊗k + |1⟩⊗k?
(b) Show that it is possible to parallelize the implementation of Ukϕ in such a way that all k of the Uϕ gates are applied in parallel (on different qubits). You may perform standard quantum gates on the qubits before and after the application of the k parallel phase gates.
2023-03-20