Table of Contents
Fetching ...

Constant-depth circuits for Boolean functions and quantum memory devices using multi-qubit gates

Jonathan Allcock, Jinge Bao, Joao F. Doriguello, Alessandro Luongo, Miklos Santha

Abstract

We explore the power of the unbounded Fan-Out gate and the Global Tunable gates generated by Ising-type Hamiltonians in constructing constant-depth quantum circuits, with particular attention to quantum memory devices. We propose two types of constant-depth constructions for implementing Uniformly Controlled Gates. These gates include the Fan-In gates defined by $|x\rangle|b\rangle\mapsto |x\rangle|b\oplus f(x)\rangle$ for $x\in\{0,1\}^n$ and $b\in\{0,1\}$, where $f$ is a Boolean function. The first of our constructions is based on computing the one-hot encoding of the control register $|x\rangle$, while the second is based on Boolean analysis and exploits different representations of $f$ such as its Fourier expansion. Via these constructions, we obtain constant-depth circuits for the quantum counterparts of read-only and read-write memory devices -- Quantum Random Access Memory (QRAM) and Quantum Random Access Gate (QRAG) -- of memory size $n$. The implementation based on one-hot encoding requires either $O(n\log^{(d)}{n}\log^{(d+1)}{n})$ ancillae and $O(n\log^{(d)}{n})$ Fan-Out gates or $O(n\log^{(d)}{n})$ ancillae and $16d-10$ Global Tunable gates, where $d$ is any positive integer and $\log^{(d)}{n} = \log\cdots \log{n}$ is the $d$-times iterated logarithm. On the other hand, the implementation based on Boolean analysis requires $8d-6$ Global Tunable gates at the expense of $O(n^{1/(1-2^{-d})})$ ancillae.

Constant-depth circuits for Boolean functions and quantum memory devices using multi-qubit gates

Abstract

We explore the power of the unbounded Fan-Out gate and the Global Tunable gates generated by Ising-type Hamiltonians in constructing constant-depth quantum circuits, with particular attention to quantum memory devices. We propose two types of constant-depth constructions for implementing Uniformly Controlled Gates. These gates include the Fan-In gates defined by for and , where is a Boolean function. The first of our constructions is based on computing the one-hot encoding of the control register , while the second is based on Boolean analysis and exploits different representations of such as its Fourier expansion. Via these constructions, we obtain constant-depth circuits for the quantum counterparts of read-only and read-write memory devices -- Quantum Random Access Memory (QRAM) and Quantum Random Access Gate (QRAG) -- of memory size . The implementation based on one-hot encoding requires either ancillae and Fan-Out gates or ancillae and Global Tunable gates, where is any positive integer and is the -times iterated logarithm. On the other hand, the implementation based on Boolean analysis requires Global Tunable gates at the expense of ancillae.
Paper Structure (28 sections, 20 theorems, 50 equations, 11 figures, 4 tables)

This paper contains 28 sections, 20 theorems, 50 equations, 11 figures, 4 tables.

Key Result

Lemma 11

A query to a $\mathsf{QRAM}$ of memory size $n$ can be simulated using $2$ queries to a $\mathsf{QRAG}$ of memory size $n$, $3$ two-qubit gates, and $1$ workspace qubit.

Figures (11)

  • Figure 1: We give constant-depth circuits for $f$-$\mathsf{UCG}$s, which contain $f$-$\mathsf{FIN}$s and a subset of quantum memory devices ($\mathsf{QMD}$) including $\mathsf{QRAM}$ (and its generalization we call $f$-$\mathsf{QRAM}$) as special cases. Although $\mathsf{QRAG}$ is not an $f$-$\mathsf{FIN}$, our (one-hot-encoding-based) construction for $\mathsf{QRAM}$ can be adapted to it.
  • Figure 2: The architecture of a Quantum Processing Unit ($\mathsf{QPU}$) with access to a quantum memory device ($\mathsf{QMD}$). The $\mathsf{QPU}$ encompasses a ($\mathop{\mathrm{poly}}\nolimits\log{n}$)-qubit input register $\mathtt{I}$ and workspace $\mathtt{W}$, a ($\log{n}$)-qubit address register $\mathtt{A}$, and an $\ell$-qubit target register $\mathtt{T}$, while the $\mathsf{QMD}$ encompasses the address register $\mathtt{A}$, the target register $\mathtt{T}$, an $n\ell$-qubit memory array $\mathtt{M}$ composed of $n$ cells $x_0,\dots,x_{n-1} \in \{0,1\}^\ell$ of $\ell$ qubits each, and a ($\mathop{\mathrm{poly}}\nolimits{n}$)-qubit auxiliary register $\mathtt{Aux}$.
  • Figure 3: (a) A sequential implementation of $f\text{-}\mathsf{UCG}_{[n]\to n}^{(n)}$. (b) The gate $|x\rangle\langle x|\otimes \mathsf{U} + \sum_{j\in\{0,1\}^n\setminus\{x\}}|j\rangle\langle j|\otimes \mathbb{I}_1$ can be implemented by employing two $\mathsf{AND}$ gates onto an ancillary qubit and controlling $\mathsf{U}\in\mathcal{U}(\mathbb{C}^{2\times 2})$ on it being in the $|1\rangle$ state barenco1995elementary.
  • Figure 4: (a) A serial circuit of unitaries $\mathsf{U}_j = e^{i\pi \alpha_j}\mathsf{Z}(\beta_j)\mathsf{H}\mathsf{Z}(\gamma_j)\mathsf{H}\mathsf{Z}(\delta_j)$ controlled on the single-qubit register $\mathtt{E}_j$ being in the $|1\rangle$ state, $j\in[n]$. (b) If the control qubits $|e\rangle_{\mathtt{E}} = \bigotimes_{j\in[n]}|e_j\rangle_{\mathtt{E}_j}$ are such that $e\in\{0,1\}^n$ has Hamming weight at most $1$ ($|e|\leq 1$), then the gates composing $\mathsf{U}_0,\dots,\mathsf{U}_{n-1}$ can be rearranged as shown in the equivalent circuit.
  • Figure 5: The circuit for an $f$-$\mathsf{UCG}^{(n)}$ using Fan-Out gates, where $f$ is a $(J,r)$-junta with $|\overline{J}| = t$. For simplicity, we include targets from $|x_i\rangle_{\mathtt{J}}$ onto all registers $\{\mathtt{J}_{z,j}\}_{z,j}$, but in reality $x_i$ is copied onto the registers $\{\mathtt{J}_{z,j}\}_j$ for all $z\in\{0,1\}^t$ such that $i\in J_z$, where $J_z$ is the set of coordinated that $f_{J|z}$ depends on. Moreover, we omit the indices in the parameters $\alpha_{z}(j)$, $\beta_{z}(j)$, $\gamma_{z}(j)$, $\delta_{z}(j)$.
  • ...and 6 more figures

Theorems & Definitions (50)

  • Definition 7: Quantum Processing Unit
  • Definition 8: Quantum Processing Unit and Quantum Memory Device
  • Definition 9: $f$-$\mathsf{QRAM}$
  • Definition 10: $\mathsf{QRAG}$
  • Lemma 11: Simulating $\mathsf{QRAM}$ with $\mathsf{QRAG}$
  • proof
  • Lemma 12: Simulating $\mathsf{QRAG}$ with $\mathsf{QRAM}$
  • proof
  • Definition 13: $f$-Uniformly Controlled Gate
  • Definition 14: $f$-Fan-In gate
  • ...and 40 more