跳到主要内容

東京大学 新領域創成科学研究科 メディカル情報生命専攻 2024年1月実施 問題11

Author

zephyr, 祭音Myyura (with the help of an anonymous contributor)

Description

A quantum state of a 1-qubit quantum computer can be represented by a 2×22 \times 2 complex matrix ρ=12(1+abicb+ic1a)\rho = \frac{1}{2} \begin{pmatrix} 1 + a & b - ic \\ b + ic & 1 - a \end{pmatrix} called a density matrix. Here, i=1i = \sqrt{-1} is the imaginary unit that satisfies i2=1i^2 = -1 and a,b,ca, b, c are real numbers that satisfy a2+b2+c21a^2 + b^2 + c^2 \leq 1. By measuring the qubit represented by matrix ρ\rho in the computational basis (also called the Z-basis), we observe state 0 or state 1 with probabilities p0=1+a2p_0 = \frac{1 + a}{2} and p1=1a2p_1 = \frac{1 - a}{2}, which are the diagonal elements of matrix ρ\rho, respectively. Answer the following questions with mathematical derivation.

  1. Show that all the eigenvalues of matrix ρ\rho are non-negative real numbers.

  2. Answer the probability that state 0 is observed by measurement in the computational basis after applying quantum gate operation H=12(1111)H = \frac{1}{\sqrt{2}} \begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix} to quantum state ρ\rho.

  3. Answer the probability that state 0 is observed by measurement in the computational basis after applying quantum gate operation Y=(0ii0)Y = \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix} to quantum state ρ\rho.

  4. Let ρ=12(1+abicb+ic1a)\rho' = \frac{1}{2} \begin{pmatrix} 1 + a' & b' - ic' \\ b' + ic' & 1 - a' \end{pmatrix} be the quantum state after applying quantum gate operation U=(u00+iu00u01+iu01u10+iu10u11+iu11)U = \begin{pmatrix} u_{00}' + iu_{00}'' & u_{01}' + iu_{01}'' \\ u_{10}' + iu_{10}'' & u_{11}' + iu_{11}'' \end{pmatrix} to quantum state ρ\rho. Compute a2+b2+c2a'^2 + b'^2 + c'^2.


1个量子比特量子计算机的量子态可以表示为一个 2×22 \times 2 的复数矩阵 ρ=12(1+abicb+ic1a)\rho = \frac{1}{2} \begin{pmatrix} 1 + a & b - ic \\ b + ic & 1 - a \end{pmatrix},称为密度矩阵。这里,i=1i = \sqrt{-1} 是满足 i2=1i^2 = -1 的虚数单位,a,b,ca, b, c 是满足 a2+b2+c21a^2 + b^2 + c^2 \leq 1 的实数。通过在计算基(也称为Z基)中测量由矩阵 ρ\rho 表示的量子比特,我们以概率 p0=1+a2p_0 = \frac{1 + a}{2}p1=1a2p_1 = \frac{1 - a}{2} 观察到状态0或状态1,分别对应密度矩阵 ρ\rho 的对角元素。用数学推导回答以下问题。

  1. 证明矩阵 ρ\rho 的所有特征值都是非负实数。

  2. 在对量子态 ρ\rho 施加量子门操作 H=12(1111)H = \frac{1}{\sqrt{2}} \begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix} 后,在计算基中测量观察到状态0的概率是多少。

  3. 在对量子态 ρ\rho 施加量子门操作 Y=(0ii0)Y = \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix} 后,在计算基中测量观察到状态0的概率是多少。

  4. ρ=12(1+abicb+ic1a)\rho' = \frac{1}{2} \begin{pmatrix} 1 + a' & b' - ic' \\ b' + ic' & 1 - a' \end{pmatrix} 是施加量子门操作 U=(u00+iu00u01+iu01u10+iu10u11+iu11)U = \begin{pmatrix} u_{00}' + iu_{00}'' & u_{01}' + iu_{01}'' \\ u_{10}' + iu_{10}'' & u_{11}' + iu_{11}'' \end{pmatrix} 后的量子态。计算 a2+b2+c2a'^2 + b'^2 + c'^2

Kai

解题思路

这道题目涉及量子计算中的密度矩阵和量子门操作,需要运用线性代数和复数运算的知识。我们将逐步解答每个小问:

  1. 求密度矩阵的特征值,证明它们是非负实数。
  2. 计算 Hadamard 门(H 门)操作后的测量概率。
  3. 计算 Y 门操作后的测量概率。
  4. 计算一般量子门 U 操作后的密度矩阵参数。

每个小问都需要详细的数学推导。

1. Show that all the eigenvalues of matrix ρ\rho are non-negative real numbers

To find the eigenvalues of ρ\rho, we need to solve the characteristic equation:

det(ρλI)=0\det(\rho - \lambda I) = 0
det((1+a2λbic2b+ic21a2λ))=0\det\left(\begin{pmatrix} \frac{1+a}{2} - \lambda & \frac{b-ic}{2} \\ \frac{b+ic}{2} & \frac{1-a}{2} - \lambda \end{pmatrix}\right) = 0
(1+a2λ)(1a2λ)(bic2)(b+ic2)=0(\frac{1+a}{2} - \lambda)(\frac{1-a}{2} - \lambda) - (\frac{b-ic}{2})(\frac{b+ic}{2}) = 0
λ2λ+1a2b2c24=0\lambda^2 - \lambda + \frac{1-a^2-b^2-c^2}{4} = 0

The solutions to this quadratic equation are:

λ=1±1(1a2b2c2)2=1±a2+b2+c22\lambda = \frac{1 \pm \sqrt{1-(1-a^2-b^2-c^2)}}{2} = \frac{1 \pm \sqrt{a^2+b^2+c^2}}{2}

Since a2+b2+c21a^2 + b^2 + c^2 \leq 1, we have 0a2+b2+c210 \leq \sqrt{a^2+b^2+c^2} \leq 1. Therefore, both eigenvalues are real and non-negative.

2. Probability of observing state 0 after applying H gate

The Hadamard gate operation transforms ρ\rho to HρHH\rho H^\dagger. Let's calculate this:

HρH=12(1111)(1+abicb+ic1a)(1111)=14((1+a)+(b+ic)+(bic)+(1a)(1+a)(b+ic)+(bic)(1a)(1+a)+(b+ic)(bic)(1a)(1+a)(b+ic)(bic)+(1a))=12(1+baica+ic1b)\begin{aligned} H\rho H^\dagger &= \frac{1}{2}\begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix} \begin{pmatrix} 1+a & b-ic \\ b+ic & 1-a \end{pmatrix} \begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix} \\ &= \frac{1}{4}\begin{pmatrix} (1+a)+(b+ic)+(b-ic)+(1-a) & (1+a)-(b+ic)+(b-ic)-(1-a) \\ (1+a)+(b+ic)-(b-ic)-(1-a) & (1+a)-(b+ic)-(b-ic)+(1-a) \end{pmatrix} \\ &= \frac{1}{2}\begin{pmatrix} 1+b & a-ic \\ a+ic & 1-b \end{pmatrix} \end{aligned}

The probability of observing state 0 is the top-left element of this matrix:

p0=1+b2p_0 = \frac{1+b}{2}

3. Probability of observing state 0 after applying Y gate

Similarly, for the Y gate:

YρY=(0ii0)(1+abicb+ic1a)(0ii0)=(1abicb+ic1+a)\begin{aligned} Y\rho Y^\dagger &= \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix} \begin{pmatrix} 1+a & b-ic \\ b+ic & 1-a \end{pmatrix} \begin{pmatrix} 0 & i \\ -i & 0 \end{pmatrix} \\ &= \begin{pmatrix} 1-a & -b-ic \\ -b+ic & 1+a \end{pmatrix} \end{aligned}

The probability of observing state 0 is:

p0=1a2p_0 = \frac{1-a}{2}

4. Compute a2+b2+c2a'^2 + b'^2 + c'^2 after applying U gate (by zephyr)

Let U=(u00+iu00u01+iu01u10+iu10u11+iu11)U = \begin{pmatrix} u_{00}' + iu_{00}'' & u_{01}' + iu_{01}'' \\ u_{10}' + iu_{10}'' & u_{11}' + iu_{11}'' \end{pmatrix} and U=(u00iu00u10iu10u01iu01u11iu11)U^\dagger = \begin{pmatrix} u_{00}' - iu_{00}'' & u_{10}' - iu_{10}'' \\ u_{01}' - iu_{01}'' & u_{11}' - iu_{11}'' \end{pmatrix}

We need to calculate UρUU\rho U^\dagger. This is a complex calculation, so let's break it down:

Uρ=((u00+iu00)(1+a)+(u01+iu01)(b+ic)(u00+iu00)(bic)+(u01+iu01)(1a)(u10+iu10)(1+a)+(u11+iu11)(b+ic)(u10+iu10)(bic)+(u11+iu11)(1a))U\rho = \begin{pmatrix} (u_{00}'+iu_{00}'')(1+a)+(u_{01}'+iu_{01}'')(b+ic) & (u_{00}'+iu_{00}'')(b-ic)+(u_{01}'+iu_{01}'')(1-a) \\ (u_{10}'+iu_{10}'')(1+a)+(u_{11}'+iu_{11}'')(b+ic) & (u_{10}'+iu_{10}'')(b-ic)+(u_{11}'+iu_{11}'')(1-a) \end{pmatrix}

Now, multiplying this by UU^\dagger and focusing on the diagonal elements:

ρ=UρU=(1+a2bic2b+ic21a2)\rho' = U\rho U^\dagger = \begin{pmatrix} \frac{1+a'}{2} & \frac{b'-ic'}{2} \\ \frac{b'+ic'}{2} & \frac{1-a'}{2} \end{pmatrix}

Where:

a=(u002u012)(1+a)+(u00u01+u00u01)2b+(u00u01u00u01)2c+(u102u112)(1a)\begin{aligned} a' &= (|u_{00}|^2-|u_{01}|^2)(1+a) + (u_{00}'u_{01}'+u_{00}''u_{01}'')2b \\ &+ (u_{00}'u_{01}''-u_{00}''u_{01}')2c + (|u_{10}|^2-|u_{11}|^2)(1-a) \end{aligned}
b=(u00u10+u00u10)(1+a)+(u01u11+u01u11)(1a)+(u00u11+u00u11+u01u10+u01u10)b(u00u11u00u11u01u10+u01u10)c\begin{aligned} b' &= (u_{00}'u_{10}'+u_{00}''u_{10}'')(1+a) + (u_{01}'u_{11}'+u_{01}''u_{11}'')(1-a) \\ &+ (u_{00}'u_{11}'+u_{00}''u_{11}''+u_{01}'u_{10}'+u_{01}''u_{10}'')b - (u_{00}'u_{11}''-u_{00}''u_{11}'-u_{01}'u_{10}''+u_{01}''u_{10}')c \end{aligned}
c=(u00u10u00u10)(1+a)+(u01u11u01u11)(1a)+(u00u11u00u11u01u10+u01u10)b+(u00u11+u00u11u01u10u01u10)c\begin{aligned} c' &= (u_{00}'u_{10}''-u_{00}''u_{10}')(1+a) + (u_{01}'u_{11}''-u_{01}''u_{11}')(1-a) \\ &+ (u_{00}'u_{11}''-u_{00}''u_{11}'-u_{01}'u_{10}''+u_{01}''u_{10}')b + (u_{00}'u_{11}'+u_{00}''u_{11}''-u_{01}'u_{10}'-u_{01}''u_{10}'')c \end{aligned}

Now, we need to compute a2+b2+c2a'^2 + b'^2 + c'^2. This is a very long and complex calculation. However, we can use a property of unitary matrices: UU=IUU^\dagger = I. This implies that the transformation ρUρU\rho \to U\rho U^\dagger preserves the trace and the purity of the density matrix.

The purity of a density matrix is defined as Tr(ρ2)=1+a2+b2+c22\text{Tr}(\rho^2) = \frac{1+a^2+b^2+c^2}{2}.

Since this quantity is preserved under unitary transformations, we have:

1+a2+b2+c22=1+a2+b2+c22\frac{1+a^2+b^2+c^2}{2} = \frac{1+a'^2+b'^2+c'^2}{2}

Therefore:

a2+b2+c2=a2+b2+c2a'^2 + b'^2 + c'^2 = a^2 + b^2 + c^2

This result shows that the sum of squares of the parameters in the density matrix is invariant under unitary transformations.

4. Compute a2+b2+c2a'^2 + b'^2 + c'^2 after applying U gate (by an anonymous contributor)

Note that

detP=1a2b2c24\det P = \frac{1 - a^2 - b^2 - c^2}{4}
detP=det(UPU)=det(PUU)=detP=1a2b2c24\det P' = \det(UPU^*) = \det(PU^*U) = \det P = \frac{1 - a^2 - b^2 - c^2}{4}

and

detP=1a2b2c24\det P' = \frac{1 - a'^2 - b'^2 - c'^2}{4}

Therefore,

a2+b2+c2=a2+b2+c2a'^2 + b'^2 + c'^2 = a^2 + b^2 + c^2

Knowledge

难点思路

第 4 小问的计算过程非常复杂,直接计算会非常繁琐。关键是要认识到酉变换的性质,即它保持密度矩阵的纯度不变。这样可以大大简化计算。

解题技巧和信息

  1. 在处理密度矩阵时,要注意其特殊性质:Hermitian(自伴)、半正定、迹为 1。
  2. 量子门操作可以表示为 UρUU\rho U^\dagger,其中 UU 是酉矩阵。
  3. 酉变换保持密度矩阵的迹和纯度不变,意味着新态的 a2+b2+c2a'^2 + b'^2 + c'^2 保持不变。这是解决复杂问题的关键。
  4. 在计算复杂的矩阵乘法时,可以先关注最终需要的元素,而不必计算整个矩阵。
  5. Hadamard 门 HH 将计算基的状态均匀地混合到对角线基。测量概率可以通过变换后的密度矩阵来计算。
  6. Pauli-Y 门 YY 交换计算基的状态并引入相位因子。

重点词汇

  • density matrix 密度矩阵
  • eigenvalue 特征值
  • quantum gate 量子门
  • Hadamard gate H 门
  • unitary transformation 酉变换
  • purity 纯度
  • trace 迹

参考资料

  1. Nielsen, M. A., & Chuang, I. L. (2010). Quantum Computation and Quantum Information: 10th Anniversary Edition. Cambridge University Press. Chapter 2 and 4.
  2. Wilde, M. M. (2017). Quantum Information Theory. Cambridge University Press. Chapter 3.