Grover Search

using Yao
using Yao.EasyBuild: variational_circuit
using LinearAlgebra

Grover Step

A single grover step is consist of applying oracle circuit and reflection circuit. The reflection_circuit function takes the wave function generator U as the input and returns U|0><0|U'.

function grover_step!(reg::AbstractRegister, oracle, U::AbstractBlock)
    apply!(reg |> oracle, reflect_circuit(U))
end

function reflect_circuit(gen::AbstractBlock)
    N = nqubits(gen)
    reflect0 = control(N, -collect(1:N-1), N=>-Z)
    chain(gen', reflect0, gen)
end

reflect_circuit (generic function with 1 method)

Compute the propotion of target states to estimate the number of iterations, which requires computing the output state.

function solution_state(oracle, gen::AbstractBlock)
    N = nqubits(gen)
    reg= zero_state(N) |> gen
    reg.state[real.(statevec(ArrayReg(ones(ComplexF64, 1<<N)) |> oracle)) .> 0] .= 0
    normalize!(reg)
end

function num_grover_step(oracle, gen::AbstractBlock)
    N = nqubits(gen)
    reg = zero_state(N) |> gen
    ratio = abs2(solution_state(oracle, gen)'*reg)
    Int(round(pi/4/sqrt(ratio)))-1
end

num_grover_step (generic function with 1 method)

Run

First, we define the problem by an oracle, it finds bit string bit"000001100100".

num_bit = 12
oracle = matblock(Diagonal((v = ones(ComplexF64, 1<<num_bit); v[Int(bit"000001100100")+1]*=-1; v)))

matblock(...)

then solve the above problem

gen = repeat(num_bit, H, 1:num_bit)
reg = zero_state(num_bit) |> gen

target_state = solution_state(oracle, gen)

for i = 1:num_grover_step(oracle, gen)
    grover_step!(reg, oracle, gen)
    overlap = abs(reg'*target_state)
    println("step $(i-1), overlap = $overlap")
end

step 0, overlap = 0.04685974121093736
step 1, overlap = 0.0780487209558483
step 2, overlap = 0.10916148124670066
step 3, overlap = 0.14016763852852288
step 4, overlap = 0.17103691335084453
step 5, overlap = 0.20173915993747182
step 6, overlap = 0.23224439562572258
step 7, overlap = 0.26252283014636996
step 8, overlap = 0.29254489471570244
step 9, overlap = 0.322281270911289
step 10, overlap = 0.35170291930325104
step 11, overlap = 0.3807811078130809
step 12, overlap = 0.40948743977231195
step 13, overlap = 0.4377938816536402
step 14, overlap = 0.46567279044741594
step 15, overlap = 0.49309694065677034
step 16, overlap = 0.5200395508850146
step 17, overlap = 0.5464743099893477
step 18, overlap = 0.5723754027753314
step 19, overlap = 0.5977175352070423
step 20, overlap = 0.6224759591082774
step 21, overlap = 0.6466264963306958
step 22, overlap = 0.6701455623652912
step 23, overlap = 0.6930101893741392
step 24, overlap = 0.7151980486199263
step 25, overlap = 0.7366874722713579
step 26, overlap = 0.7574574745631494
step 27, overlap = 0.7774877722899375
step 28, overlap = 0.7967588046140988
step 29, overlap = 0.8152517521681291
step 30, overlap = 0.8329485554329328
step 31, overlap = 0.8498319323740713
step 32, overlap = 0.8658853953187506
step 33, overlap = 0.8810932670570639
step 34, overlap = 0.8954406961517668
step 35, overlap = 0.9089136714416339
step 36, overlap = 0.9214990357242339
step 37, overlap = 0.9331844986047592
step 38, overlap = 0.9439586484983656
step 39, overlap = 0.953810963774298
step 40, overlap = 0.9627318230309194
step 41, overlap = 0.9707125144916121
step 42, overlap = 0.9777452445123718
step 43, overlap = 0.983823145192787
step 44, overlap = 0.9889402810829753
step 45, overlap = 0.9930916549799182
step 46, overlap = 0.9962732128075449
step 47, overlap = 0.9984818475757891
step 48, overlap = 0.9997154024147601

Rejection Sampling

In practise, it is often not possible to determine the number of iterations before actual running. we can use rejection sampling technique to avoid estimating the number of grover steps.

In a single try, we apply the grover algorithm for nstep times.

function single_try(oracle, gen::AbstractBlock, nstep::Int; nbatch::Int)
    N = nqubits(gen)
    reg = zero_state(N+1; nbatch)
    focus(reg, (1:N...,)) do r
        r |> gen
        for i = 1:nstep
            grover_step!(r, oracle, gen)
        end
        return r
    end
    reg |> checker
    res = measure!(RemoveMeasured(), reg, (N+1))
    return res, reg
end

single_try (generic function with 1 method)

After running the grover search, we have a checker program that flips the ancilla qubit if the output is the desired value, we assume the checker program can be implemented in polynomial time. to gaurante the output is correct. We contruct a checker "program", if the result is correct, flip the ancilla qubit

ctrl = -collect(1:num_bit); ctrl[[3,6,7]] *= -1
checker = control(num_bit+1,ctrl, num_bit+1=>X)

nqubits: 13
control(¬1, ¬2, 3, ¬4, ¬5, 6, 7, ¬8, ¬9, ¬10, ¬11, ¬12)
└─ (13,) X

The register is batched, with batch dimension nshot. focus! views the first 1-N qubts as system. For a batched register, measure! returns a vector of bitstring as output.

Run

maxtry = 100
nshot = 3

for nstep = 0:maxtry
    println("number of iter = $nstep")
    res, regi = single_try(oracle, gen, nstep; nbatch=3)

    # success!
    if any(==(1), res)
        overlap_final = viewbatch(regi, findfirst(==(1), res))'*target_state
        println("success, overlap = $(overlap_final)")
        break
    end
end

number of iter = 0
number of iter = 1
number of iter = 2
number of iter = 3
number of iter = 4
number of iter = 5
number of iter = 6
number of iter = 7
success, overlap = -1.0 + 0.0im

The final state has an overlap of 1 with the target state.

Amplitude Amplification

Given a circuit to generate a state, now we want to project out the subspace with [1,3,5,8,9,11,12] fixed to 1 and [4,6] fixed to 0. We can construct an oracle

evidense = [1, 3, -4, 5, -6, 8, 9, 11, 12]
function inference_oracle(nbit::Int, locs::Vector{Int})
    control(nbit, locs[1:end-1], abs(locs[end]) => (locs[end]>0 ? Z : -Z))
end
oracle = inference_oracle(nqubits(reg), evidense)

nqubits: 12
control(1, 3, ¬4, 5, ¬6, 8, 9, 11)
└─ (12,) Z

We use a variational circuit generator defined in Yao.EasyBuild

gen = dispatch!(variational_circuit(num_bit), :random)
reg = zero_state(num_bit) |> gen

ArrayReg{2, ComplexF64, Array...}
    active qubits: 12/12
    nlevel: 2

Run

solution = solution_state(oracle, gen)
for i = 1:num_grover_step(oracle, gen)
    grover_step!(reg, oracle, gen)
    println("step $(i-1), overlap = $(abs(reg'*solution))")
end

step 0, overlap = 0.11480007965675817
step 1, overlap = 0.19058324193670428
step 2, overlap = 0.2652457005602443
step 3, overlap = 0.33834841120466996
step 4, overlap = 0.40946150147259774
step 5, overlap = 0.4781667987090151
step 6, overlap = 0.5440602890192854
step 7, overlap = 0.6067544930281308
step 8, overlap = 0.6658807444092039
step 9, overlap = 0.7210913577866083
step 10, overlap = 0.7720616732602615
step 11, overlap = 0.8184919655324984
step 12, overlap = 0.8601092064095035
step 13, overlap = 0.89666867031338
step 14, overlap = 0.9279553733638201
step 15, overlap = 0.9537853375670116
step 16, overlap = 0.974006672677868
step 17, overlap = 0.9885004693737958
step 18, overlap = 0.997181498487807

This page was generated using Literate.jl.