Neutral atom quantum circuit compiling method and system, electronic equipment, storage medium and computer program product

Abstract

The invention discloses a neutral atom quantum circuit compiling method and system, electronic equipment, a storage medium and a computer program product. Comprising the following steps: constructing a two-bit-gate frequency diagram based on a two-bit-gate quantum circuit in a two-bit-gate quantum circuit; determining Hamiltonian of simulation quantum calculation; coding the two-bit gate frequency diagram based on a maximum cut problem algorithm to obtain a maximum loss function corresponding to the two-bit gate frequency diagram; adjusting the laser intensity of the optical tweezer array and the atomic distance between neutral atoms in the optical tweezer array to obtain a target Hamiltonian meeting a preset condition; solving the maximum loss function based on the target Hamiltonian to obtain two groups of corresponding solutions; performing segmentation processing on the two-bit gate frequency diagram based on the two groups of corresponding solutions to obtain two quantum bit groups; and performing two bit gates on the two quantum bit groups between the groups and in the groups based on the optical tweezers array until all the quantum bits are executed.

Description

Technical Field

[0001] The present invention relates to the technical field of quantum computing, and in particular to a neutral atom quantum circuit compilation method and system, electronic equipment, storage medium, and computer program product. Background Art

[0002] A neutral atom is an atom with no net electrical charge. The number of protons in its nucleus is equal to the number of electrons outside the nucleus, and the positive and negative charges cancel each other out, making the atom electrically neutral overall. In quantum computing, neutral atom quantum computing uses neutral atoms (such as alkali metal atoms like rubidium (Rb) and cesium (Cs)) as the physical carriers of quantum bits (qubits). These qubits are encoded and manipulated using optical tweezers arrays and laser pulses, enabling quantum computing tasks.

[0003] Neutral atom quantum circuits, in neutral atom quantum computing, transform quantum algorithms or quantum operations into executable processes consisting of a series of physical steps (such as laser pulses and Rydberg excitations) to manipulate neutral atoms. These operations operate on neutral atom qubits, and quantum computing tasks are achieved through combinations of quantum gates. Compiling neutral atom quantum circuits requires translating abstract quantum algorithms into physical operations executable by actual hardware, involving quantum gate decomposition, layout and routing, and error correction coding.

[0004] The characteristic of neutral atoms is that, under the constraints of the optical tweezers array, the neutral atoms bound in the optical tweezers array can move their positions coherently, so that the connectivity between quantum bits will change in real time during the quantum computing process.

[0005] The inventors discovered that the mobility of neutral atoms allows for greater flexibility in the application of two-qubit gates in quantum computing. However, the dynamic connectivity and enormous amounts of quantum data complicate the compilation of neutral atom quantum circuits. The present invention addresses the technical challenge of rapidly compiling such a large number of qubits, leveraging the unique properties of neutral atom quantum computing (such as coherent mobility).

[0006] The contents of the background technology section are merely technologies known to the public and do not necessarily represent the existing technologies in this field. Summary of the Invention

[0007] According to one aspect of the present invention, the present invention provides a neutral atom quantum circuit compilation method, including mapping the quantum bits of a preset quantum circuit with the neutral atoms in an optical tweezers array to obtain a two-bit gate quantum circuit; constructing a two-bit gate frequency map based on the two-bit gate quantum circuit in the two-bit gate quantum circuit; determining the Hamiltonian for simulated quantum computing; encoding the two-bit gate frequency map based on a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-bit gate frequency map; adjusting the laser intensity of the optical tweezers array and the atomic distance between the neutral atoms in the optical tweezers array to obtain a target Hamiltonian that meets preset conditions; solving the maximized loss function based on the target Hamiltonian to obtain two groups of corresponding solutions; segmenting the two-bit gate frequency map based on the two groups of corresponding solutions to obtain two quantum bit groups; and executing two-bit gates between and within the two groups of quantum bit groups based on the optical tweezers array until all quantum bits are executed.

[0008] According to another aspect of the present invention, a neutral atom quantum circuit compilation system is provided, comprising a quantum circuit processing module, a qubit processing module, and a quantum circuit compilation module. The quantum circuit processing module maps the qubits of a preset quantum circuit with the neutral atoms in an optical tweezers array to obtain a two-qubit gate quantum circuit, and constructs a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit. The qubit processing module determines the Hamiltonian for simulating quantum computation, encodes the two-qubit gate frequency map using a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-qubit gate frequency map, adjusts the laser intensity of the optical tweezers array and the atomic distances between the neutral atoms in the optical tweezers array to obtain a target Hamiltonian that meets preset conditions, solves the maximized loss function based on the target Hamiltonian to obtain two corresponding solutions, and segments the two-qubit gate frequency map based on the two corresponding solutions to obtain two qubit groups. The quantum circuit compilation module executes two-qubit gates on the two qubit groups, both within and between the groups, using the optical tweezers array, until all qubits are executed.

[0009] According to yet another aspect of the present invention, an electronic device is provided. The electronic device includes: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method described above.

[0010] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method described above can be implemented.

[0011] According to another aspect of the present invention, a computer program product is provided, which includes a computer program stored on a computer-readable storage medium; the computer program includes program instructions, which, when executed by a computer, cause the computer to execute the method described above.

[0012] Beneficial effects

[0013] The present invention maps the qubits of a preset quantum circuit to neutral atoms in an optical tweezers array to obtain a two-qubit gate quantum circuit, and constructs a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit. The present invention can determine the Hamiltonian for simulated quantum computing, encode the two-qubit gate frequency map based on a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-qubit gate frequency map, and obtain a target Hamiltonian that meets preset conditions by adjusting the laser intensity of the optical tweezers array and the atomic distances between neutral atoms in the optical tweezers array. The maximized loss function is then solved based on the target Hamiltonian to obtain two corresponding solutions. The two-qubit gate frequency map is segmented based on the two corresponding solutions to obtain two qubit groups. Finally, two-qubit gates can be executed on the two qubit groups, both within and between the groups, using the optical tweezers array, until all qubits are executed.

[0014] By leveraging the neutral atom quantum computing system's capabilities, both for analog and digital quantum computing, this invention can complete the entire quantum circuit compilation process using only quantum computing power. By simulating quantum computing's ability to solve the maximum cut problem algorithm, this invention can accelerate the computational speed of the circuit compilation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A schematic diagram illustrating a flow chart of a neutral atom quantum circuit compilation method according to an embodiment of the present invention;

[0017] Figure 2 A schematic diagram showing an SLM optical tweezers array according to an embodiment of the present invention;

[0018] Figure 3 A schematic diagram showing an AOD optical tweezers array according to an embodiment of the present invention;

[0019] Figure 4 A schematic diagram showing a two-qubit gate quantum circuit according to an embodiment of the present invention is shown;

[0020] Figure 5 A two-bit gate frequency diagram is shown according to an embodiment of the present invention;

[0021] Figure 6 A schematic diagram showing a cut diagram of a two-bit gate frequency diagram according to an embodiment of the present invention;

[0022] Figure 7 Another schematic diagram illustrating a flow chart of a neutral atom quantum circuit compilation method according to an embodiment of the present invention;

[0023] Figure 8 Another schematic diagram illustrating a flow chart of a neutral atom quantum circuit compilation method according to an embodiment of the present invention;

[0024] Figure 9 A schematic diagram showing a neutral atom quantum circuit compilation system according to an embodiment of the present invention.

[0025] Description of reference numerals:

[0026] Neutral atom quantum circuit compilation system 1; quantum circuit processing module 10; quantum bit processing module 20; quantum circuit compilation module 30. DETAILED DESCRIPTION

[0027] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of the present invention.

[0028] In one prior art, a mobile neutral atom quantum circuit can be compiled using a quantum circuit compiler. For example, an SMT (Satisfiability Modulo Theories) solver can be used to solve the mapping and scheduling process of quantum bits that meet physical constraints.

[0029] However, the inventors discovered that the scalability of this compilation method is limited by the SMT solver, causing the solution time to increase exponentially with the number of quantum circuits, and the computational resources required to complete the solution are high. For example, for a quantum circuit with 90 qubits, it would take at least one day to complete the compilation.

[0030] In another prior art, a movable neutral atom quantum circuit can be compiled based on the MAX k-Cut algorithm. For example, qubits are mapped based on the MAX k-Cut algorithm and the cut qubits are placed in a K-set optical tweezers array to remove the limitations of neutral atom quantum computing.

[0031] However, the inventors discovered that this compilation method requires at least K-1 sets of AODs to constrain qubits. Current neutral atom quantum computing platforms are limited to two AODs, making multiple AOD deployments difficult to implement in practical applications. Furthermore, the MAX k-Cut algorithm is an NP-hard problem, posing significant computational resource constraints and hindering rapid compilation and solution for large-scale neutral atom quantum computing.

[0032] In another prior art, corresponding solutions and integration of these solutions are proposed for the three tasks in DPQA (Discrete-Pulse Quantum Assembly, quantum circuit compiler), such as scheduling tasks, layout tasks, and routing tasks, based on the edge coloring problem, simulated annealing problem, and independent set problem.

[0033] However, the inventors found that the compiling method is limited to circuits that can commutate two-bit gate groups, lacking broad adaptability. Furthermore, the compiling method only considers the shortest distance issue while ignoring the compatibility issue during routing, resulting in an inability to achieve optimal compilation.

[0034] In another prior art, all qubits can be mapped to fixed atoms, and then movable atoms are used as auxiliary atoms to perform routing between the fixed qubits and perform 2-Q gates.

[0035] However, the inventors found that this compilation method transmits two-bit gate interactions through auxiliary bits, and additional two-bit gates need to be added in this process to complete it. This will increase the number of two-bit gate operations required for the entire quantum circuit, thereby reducing the fidelity of the entire quantum circuit.

[0036] That is, the compilation of neutral atom quantum circuits in the existing technology has problems such as huge computing resources, long compilation time, inability to obtain optimal compilation, and low fidelity of quantum circuits.

[0037] Based on this, according to one aspect of the present invention, the present invention provides a neutral atom quantum circuit compilation method. Figure 1 A schematic flow chart of a neutral atom quantum circuit compilation method according to an embodiment of the present invention is shown. Figure 1As shown, the neutral atom quantum circuit compilation method may include steps S100-S800.

[0038] Exemplarily, the neutral atom quantum circuit compilation method can be executed by a neutral atom quantum circuit compilation system (or neutral atom quantum circuit computing platform) with computing capabilities.

[0039] According to an example embodiment, in step S100 , the neutral atom quantum circuit compilation system maps the quantum bits of a preset quantum circuit with the neutral atoms in the optical tweezers array to obtain a two-bit gate quantum circuit.

[0040] For example, the preset quantum circuit can be a quantum circuit task input to a neutral atom quantum circuit compilation system. The neutral atom quantum circuit compilation system can use neutral atoms bound in an optical tweezers array as mapping targets to construct a mapping between the qubits of the preset quantum circuit and the neutral atoms. For example, the optical tweezers array can include an SLM optical tweezers array and an AOD optical tweezers array.

[0041] An optical tweezers array is a quantum manipulation platform that can capture and manipulate atoms or molecules using laser beams. The array can include multiple optical tweezers sites, each of which can accommodate at most one ground-state atom.

[0042] Optical tweezers sites can be generated using either an SLM (Spatial Light Modulator) or an AOD (Acousto-optic Deflector). An optical tweezers array formed by sites generated by an SLM is called an SLM array; an optical tweezers array formed by sites generated by an AOD is called an AOD array.

[0043] Figure 2 A schematic diagram showing an SLM optical tweezers array according to an embodiment of the present invention; Figure 3 A schematic diagram of an AOD optical tweezers array according to an embodiment of the present invention is shown.

[0044] like Figure 2 As shown, the optical tweezers points in the SLM optical tweezers array can be at any position on the two-dimensional plane, but the optical tweezers points in the SLM optical tweezers array cannot be moved. For example, Figure 2 An SLM optical tweezers array comprising seven optical tweezers points is shown, and the positions of these seven optical tweezers points are randomly distributed.

[0045] like Figure 3 As shown in FIG, the optical tweezers points in the AOD optical tweezers array are a lattice array structure located on a two-dimensional plane. The AOD optical tweezers array can be moved, but the AOD optical tweezers array needs to move in a whole row or column during the movement. For example, Figure 3As shown, the row 3 in the AOD optical tweezers array can be moved downward as a whole. The movement of the AOD optical tweezers array can drive the position of the neutral atoms to move.

[0046] According to example embodiments, neutral atoms can be switched between an SLM optical tweezers array and an AOD optical tweezers array. The optical tweezers points of the SLM optical tweezers array can be individually turned on or off during quantum computing, while the optical tweezers points of the AOD optical tweezers array need to be turned on or off as a whole row or column during quantum computing.

[0047] Neutral atoms in the AOD optical tweezers array can only move sequentially. That is, a row of the AOD optical tweezers array can only move sequentially in row order and cannot directly move across the middle row. Similarly, a column of the AOD optical tweezers array can only move sequentially in column order and cannot directly move across the middle column.

[0048] According to example embodiments, any single-qubit gate can be individually controlled by lasers. It is understood that when the distance between two qubits is within the Rydberg blockade radius r b When the two qubits execute the CZ gate, the distance between the two neutral atoms (r b ) cannot contain other neutral atoms.

[0049] For example, in step S100, the neutral atom quantum circuit compilation system converts any two-bit gate in the original circuit into a CZ gate that can be natively executed by neutral atoms, thereby obtaining a two-bit gate quantum circuit with only two-bit gates.

[0050] In step S200 , the neutral atom quantum circuit compilation system constructs a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit.

[0051] For example, the neutral atom quantum circuit compilation system can determine the two-bit gate quantum circuit in a two-bit gate quantum circuit. Based on the two-bit gate quantum circuit, the number of times the two-bit gate between two quantum bits appears in the entire quantum circuit can be determined, and a two-bit gate frequency map can be constructed accordingly.

[0052] Therefore, the two-bit gate frequency map can be determined based on the number of times the two-bit gate between two qubits appears in the entire quantum circuit.

[0053] Figure 4 A schematic diagram showing a two-qubit gate quantum circuit according to an embodiment of the present invention is shown; Figure 5A two-bit gate frequency diagram according to an embodiment of the present invention is shown.

[0054] For example, Figure 4 A schematic diagram of a two-bit gate quantum circuit including five quantum bits (q1, q2, q3, q4, q5) is shown. Figure 4 The two-bit gate quantum circuit shown can be determined as follows Figure 5 The two-bit gate frequency diagram is shown.

[0055] In step S300 , the neutral atom quantum circuit compilation system determines a Hamiltonian for simulating quantum computation.

[0056] For example, in a neutral atom quantum circuit compilation system, neutral atoms can be used for both simulated quantum computing and digital quantum computing. Simulated quantum computing directly uses the natural evolution of quantum computing to simulate specific problems without decomposing the calculation into discrete logic gate operations. It can solve problems by regulating the Hamiltonian of the simulated quantum computing (such as interactions, magnetic fields, etc.) so that the quantum computing naturally evolves to the target state. Mathematical quantum computing is a universal computing model based on quantum logic gates. It implements arbitrary quantum algorithms through discrete gate operations. It can decompose the calculation into a combination of a series of universal quantum gates (such as Hadamard gates, CNOT gates, etc.) and implement the algorithm through quantum circuits.

[0057] In simulated quantum computing, the ground state of a neutral atom is the zero state, and the Rydberg state of a neutral atom is the one state. During the compilation of a neutral atom quantum circuit, single-bit operations (or single-bit rotations) can be achieved by controlling the laser field of the optical tweezers array, which can change the phase and amplitude of the neutral atom's internal state. Neutral atoms have a high principal quantum number and strong dipole interaction in the Rydberg state, which can be used for the interaction and control between adjacent quantum bits in quantum computing. For example, because neutral atoms have a strong dipole moment in the Rydberg state, multiple neutral atoms can be excited into the Rydberg state by laser, and arbitrary multi-atom interactions can be achieved by adjusting the positions of the neutral atoms.

[0058] In step S300 , the neutral atom quantum circuit compilation system may determine a Hamiltonian for simulating quantum computing based on neutral atoms.

[0059] Optionally, the neutral atom quantum circuit compilation system determines the Hamiltonian for simulated quantum computing based on the ground state or Rydberg state of the neutral atoms and the interaction strength between neutral atoms.

[0060] For example, the Hamiltonian for simulating quantum computing can be:

[0061]

[0062] Where H is the Hamiltonian, Ω is the Rabi driving frequency between the 0 state and the 1 state, and the magnitude of the Rabi driving frequency determines the time required for the system to go back and forth between the two states. i is the 0 state of the i-th neutral atom, r i is the 1 state of the i-th neutral atom, Δ j is the detuning amount on the jth neutral atom (i.e., the energy felt by the jth neutral atom at this position), n j =|r j > <r j |,n j Indicates whether the jth neutral atom is in state 1, n k =|g k > <g k |,n k Indicates whether the kth neutral atom is in the 0 state, V jk is the interaction strength between the jth neutral atom and the kth neutral atom.

[0063] In step S400 , the neutral atom quantum circuit compilation system encodes the two-bit gate frequency map based on the maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-bit gate frequency map.

[0064] For example, the maximum cut problem algorithm can be a Max-Cut problem algorithm. The Max-Cut problem algorithm is a classic optimization problem in graph theory. The goal is to partition a vertex set into two subsets in an undirected graph so that the sum of the weights of the edges connecting the two subsets is maximized.

[0065] Optionally, in step S400 , the neutral atom quantum circuit compilation system determines a maximized loss function based on the number of two-bit gate operations between neutral atoms according to a maximum cut problem algorithm.

[0066] For example, the maximized loss function corresponding to the two-bit gate frequency map can be:

[0067]

[0068] Among them, C Max-Cut To maximize the loss function, ω jk is the weight of the line between the jth neutral atom and the kth neutral atom, that is, ω jk is the number of two-bit gate operations between the j-th neutral atom and the k-th neutral atom, X j is the maximum cut component corresponding to the jth neutral atom, X j is 0 or 1, indicating that the jth neutral atom is in component 0 or 1, X k is the maximum cut component corresponding to the kth neutral atom, X kis 0 or 1, indicating that the jth neutral atom is in component 0 or 1, and E is the number of neutral atoms.

[0069] In step S500 , the neutral atom quantum circuit compilation system adjusts the laser intensity of the optical tweezers array and the atomic distances between neutral atoms in the optical tweezers array to obtain a target Hamiltonian that meets preset conditions.

[0070] For example, the neutral atom quantum circuit compilation system can respond to user instructions and adjust the laser intensity of the optical tweezers array and the atomic distance between the neutral atoms in the optical tweezers array to make Ω, V jk and Δ j By satisfying specific conditions and achieving the required target Hamiltonian, the solution to the Max-Cut problem is given. For example, by modulating the laser intensity, the equivalent Rabi drive frequency Ω can be achieved, and by adjusting the laser resonant frequency, the detuning amount Δ can be adjusted. j By changing the distance between the two neutral atoms through the optical tweezers array, the interaction strength V between the two neutral atoms can be adjusted. jk , thereby changing the coupling strength of the two quantum bits.

[0071] If Ω, V jk and Δ j The following conditions are met:

[0072]

[0073] Substituting the above parameters into the Hamiltonian, the target Hamiltonian is:

[0074]

[0075] set up:

[0076]

[0077] Then the target Hamiltonian is obtained as:

[0078]

[0079] In step S600 , the neutral atom quantum circuit compilation system solves the maximized loss function based on the target Hamiltonian to obtain two sets of corresponding solutions.

[0080] Optionally, in step S600 , the neutral atom quantum circuit compilation system measures the ground state of the simulated quantum computation based on the target Hamiltonian to obtain two sets of corresponding solutions.

[0081] For example, suppose:

[0082]

[0083] Among them, Z j and Z k The value of is only ±1.

[0084] Substituting the above parameters into the maximum loss function, we get:

[0085] C Max-Cut =∑ j ∑ k ω jk (1-Z j Z k );

[0086] The maximum loss function is solved based on the above parameters. For example, due to the jk If the individual sums are constant, maximizing the loss function requires minimizing the second sum, meaning the solution is consistent with the ground state of the target Hamiltonian. In other words, given the aforementioned parameters, the neutral atom quantum circuit compilation system can simulate the ground state of quantum computation based on the target Hamiltonian, yielding two corresponding solutions that maximize the loss function.

[0087] In step S700 , the neutral atom quantum circuit compilation system divides the two-qubit gate frequency map based on the two groups of corresponding solutions to obtain two qubit groups.

[0088] For example, the neutral atom quantum circuit compilation system can split the neutral atoms into two groups with the maximum number of connections based on the two corresponding solutions that maximize the loss function, thereby obtaining a cutting graph of the two-bit gate frequency graph.

[0089] Figure 6 A schematic diagram illustrating a cut diagram of a two-bit gate frequency diagram according to an embodiment of the present invention.

[0090] like Figure 6 As shown, the cutting graph includes two quantum bit groups, such as quantum bit groups {q2, q3} and {q1, q4, q5}.

[0091] In step S800, the neutral atom quantum circuit compilation system executes two-bit gates between and within two groups of quantum bits based on the optical tweezers array until all quantum bits are executed.

[0092] For example, the neutral atom quantum circuit compilation system can realize two-bit gates between quantum bits based on the coherent movement control of neutral atoms by an optical tweezers array.

[0093] For example, the neutral atom quantum circuit compilation system can first execute a two-bit gate of qubits between two qubit groups based on the optical tweezers array, and then execute a two-bit gate of qubits within a qubit group based on the optical tweezers array.

[0094] Optionally, the optical tweezers array includes an SLM optical tweezers array and an AOD optical tweezers array.

[0095] Figure 7 Another schematic diagram illustrating a flow chart of a neutral atom quantum circuit compilation method according to an embodiment of the present invention; Figure 8 Another flowchart of the neutral atom quantum circuit compilation method according to an embodiment of the present invention is shown.

[0096] like Figure 7 As shown, step S800 may include steps S810 and S820.

[0097] In step S810, the neutral atom quantum circuit compilation system moves the neutral atoms set in the AOD optical tweezers array that are mapped to the first group of quantum bits to the position of the neutral atoms set in the SLM optical tweezers array that are mapped to the second group of quantum bits, so that a two-bit gate is executed between the first group of quantum bits and the second group of quantum bits.

[0098] For example, when the neutral atom quantum circuit compilation system executes a two-qubit gate between groups, the qubits in the first qubit group are mapped to neutral atoms arranged in the SLM optical tweezers array, and the qubits in the second qubit group are mapped to neutral atoms arranged in the AOD optical tweezers array. The neutral atom quantum circuit compilation system can control the movement of the neutral atoms in the AOD optical tweezers array through the AOD optical tweezers array, thereby moving the neutral atoms in the AOD optical tweezers array to the neutral atoms in the SLM optical tweezers array. This arrangement allows the two-qubit gate between the two qubit groups to be executed in a maximally parallel manner.

[0099] Alternatively, as Figure 8 As shown, step S820 may further include steps S821-S824.

[0100] In step S821 , the neutral atom quantum circuit compilation system determines a target neutral atom mapped to the target qubits in the first and second qubit groups and disposed in the SLM optical tweezers array.

[0101] For example, after executing a two-bit gate between the first qubit group and the second qubit group, all neutral atoms can be set in the SLM optical tweezers array so that the target qubits in the first qubit group and the second qubit group can be mapped to the target neutral atoms of the SLM optical tweezers array.

[0102] In step S822 , the neutral atom quantum circuit compilation system moves the target neutral atom to the target position in the AOD optical tweezers array based on the AOD optical tweezers array, so that the target qubit and the qubit mapped by the neutral atom at the target position execute a two-bit gate.

[0103] For example, the neutral atom quantum circuit compilation system based on the AOD optical tweezers array can tweeze one target neutral atom set in the SLM optical tweezers array to the position of another neutral atom at a time, so that the mapped quantum bit can execute a two-bit gate.

[0104] In step S823 , after executing the two-bit gate, the neutral atom quantum circuit compilation system moves the target neutral atom back to its initial position based on the AOD optical tweezers array.

[0105] In step S824, the neutral atom quantum circuit compilation system traverses all target neutral atoms mapped by the target qubits in the first and second qubit groups and set in the SLM optical tweezers array until the two-bit gates of all target qubits are executed.

[0106] For example, after executing a two-qubit gate for a target qubit, the neutral atom quantum circuit compilation system uses the AOD optical tweezers array to move the corresponding neutral atom back to its initial position within the SLM optical tweezers array. This process is repeated until all two-qubit gates have been executed, completing the entire quantum circuit and achieving neutral atom quantum circuit compilation.

[0107] The present invention maps the qubits of a preset quantum circuit to neutral atoms in an optical tweezers array to obtain a two-qubit gate quantum circuit, and constructs a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit. The present invention can determine the Hamiltonian for simulated quantum computing, encode the two-qubit gate frequency map based on a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-qubit gate frequency map, and obtain a target Hamiltonian that meets preset conditions by adjusting the laser intensity of the optical tweezers array and the atomic distances between neutral atoms in the optical tweezers array. The maximized loss function is then solved based on the target Hamiltonian to obtain two corresponding solutions. The two-qubit gate frequency map is segmented based on the two corresponding solutions to obtain two qubit groups. Finally, two-qubit gates can be executed on the two qubit groups, both within and between the groups, using the optical tweezers array, until all qubits are executed.

[0108] Through the above-described embodiments, the present invention utilizes the characteristics of both simulated and digital quantum computing in a neutral atom quantum computing system, enabling the entire quantum circuit compilation process to be completed using only quantum computing power. By simulating quantum computing's ability to solve the maximum cut problem algorithm, the present invention can accelerate the computational speed of the circuit compilation process.

[0109] The present invention can combine analog quantum computing and digital quantum computing to compile neutral atom quantum circuits. No classical computing power is required to implement the entire quantum algorithm. Based on analog quantum computing, some NP problems (such as the maximum cut problem) can be quickly solved. The quantum algorithm is then compiled through digital quantum computing, and the compilation process can be mapped to the maximum cut problem, allowing the solution of the mapping scheme to be completed using analog quantum computing. Furthermore, through the routing design of neutral atoms, the present invention can enable the mapping scheme obtained from analog quantum computing to be run on the digital quantum computing platform, thereby completing the rapid compilation of quantum circuits.

[0110] The present invention can implement the neutral atom quantum circuit compilation method described above by only using one set of SLM equipment and one set of AOD equipment, does not require additional physical equipment restrictions, and has the characteristic of simple structure.

[0111] According to yet another aspect of the present invention, the present invention provides a neutral atom quantum circuit compilation system. Figure 9 FIG. 1 is a schematic diagram showing a neutral atom quantum circuit compilation system according to an embodiment of the present invention. Figure 9 As shown, the neutral atom quantum circuit compilation system 1 includes a quantum circuit processing module 10 , a quantum bit processing module 20 and a quantum circuit compilation module 30 .

[0112] According to an example embodiment, the quantum circuit processing module 10 maps the quantum bits of a preset quantum circuit with neutral atoms in the optical tweezers array to obtain a two-bit gate quantum circuit.

[0113] For example, the preset quantum circuit can be a quantum circuit task input to the quantum circuit processing module 10. The quantum circuit processing module 10 can use neutral atoms bound in an optical tweezers array as mapping targets to construct a mapping between the quantum bits of the preset quantum circuit and the neutral atoms. For example, the optical tweezers array can include an SLM optical tweezers array and an AOD optical tweezers array.

[0114] The optical tweezers array has been introduced in detail above and will not be described again here.

[0115] According to example embodiments, any single-qubit gate can be individually controlled by lasers. It is understood that when the distance between two qubits is within the Rydberg blockade radius r bWhen the two qubits execute the CZ gate, the distance between the two neutral atoms (r b ) cannot contain other neutral atoms.

[0116] For example, the quantum circuit processing module 10 converts any two-bit gate in the original circuit into a CZ gate that can be natively executed by neutral atoms, thereby obtaining a two-bit gate quantum circuit with only two-bit gates.

[0117] According to an example embodiment, the quantum circuit processing module 10 constructs a two-qubit gate frequency map based on a two-qubit gate quantum circuit in a two-qubit gate quantum circuit.

[0118] For example, the quantum circuit processing module 10 can determine a two-bit gate quantum circuit in a two-bit gate quantum circuit, and based on the two-bit gate quantum circuit, determine the number of times a two-bit gate between two quantum bits appears in the entire quantum circuit, and construct a two-bit gate frequency map accordingly.

[0119] According to an example embodiment, qubit processing module 20 determines a Hamiltonian for a simulated quantum computation.

[0120] For example, in a neutral atom quantum circuit compilation system, neutral atoms can be used for both simulated quantum computing and digital quantum computing. Simulated quantum computing directly uses the natural evolution of quantum computing to simulate specific problems without decomposing the calculation into discrete logic gate operations. It can solve problems by regulating the Hamiltonian of the simulated quantum computing (such as interactions, magnetic fields, etc.) so that the quantum computing naturally evolves to the target state. Mathematical quantum computing is a universal computing model based on quantum logic gates. It implements arbitrary quantum algorithms through discrete gate operations. It can decompose the calculation into a combination of a series of universal quantum gates (such as Hadamard gates, CNOT gates, etc.) and implement the algorithm through quantum circuits.

[0121] In simulated quantum computing, the ground state of a neutral atom is the zero state, and the Rydberg state of a neutral atom is the one state. During the compilation of a neutral atom quantum circuit, single-bit operations (or single-bit rotations) can be achieved by controlling the laser field of the optical tweezers array, which can change the phase and amplitude of the neutral atom's internal state. Neutral atoms have a high principal quantum number and strong dipole interaction in the Rydberg state, which can be used for the interaction and control between adjacent quantum bits in quantum computing. For example, because neutral atoms have a strong dipole moment in the Rydberg state, multiple neutral atoms can be excited into the Rydberg state by laser, and arbitrary multi-atom interactions can be achieved by adjusting the positions of the neutral atoms.

[0122] The qubit processing module 20 can determine the Hamiltonian for simulating quantum computing based on neutral atoms.

[0123] Optionally, the qubit processing module 20 determines the Hamiltonian for simulating quantum computing based on the ground state or Rydberg state of the neutral atoms and the interaction strength between the neutral atoms.

[0124] For example, the Hamiltonian for simulating quantum computing can be:

[0125]

[0126] Where H is the Hamiltonian, Ω is the Rabi driving frequency between the 0 state and the 1 state, and the magnitude of the Rabi driving frequency determines the time required for the system to go back and forth between the two states. i is the 0 state of the i-th neutral atom, r i is the 1 state of the i-th neutral atom, Δ j is the detuning amount on the jth neutral atom (i.e., the energy felt by the jth neutral atom at this position), n j =|r j > <r j |,n j Indicates whether the jth neutral atom is in state 1, n k =|g k > <g k |,n k Indicates whether the kth neutral atom is in the 0 state, V jk is the interaction strength between the jth neutral atom and the kth neutral atom.

[0127] The qubit processing module 20 encodes the two-bit gate frequency map based on the maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-bit gate frequency map.

[0128] For example, the maximum cut problem algorithm can be a Max-Cut problem algorithm. The Max-Cut problem algorithm is a classic optimization problem in graph theory. The goal is to partition a vertex set into two subsets in an undirected graph so that the sum of the weights of the edges connecting the two subsets is maximized.

[0129] The qubit processing module 20 determines a maximized loss function based on the number of two-bit gate operations between neutral atoms according to a maximum cut problem algorithm.

[0130] For example, the maximized loss function corresponding to the two-bit gate frequency map can be:

[0131]

[0132] Among them, C Max-CutTo maximize the loss function, ω jk is the weight of the line between the jth neutral atom and the kth neutral atom, that is, ω jk is the number of two-bit gate operations between the j-th neutral atom and the k-th neutral atom, X j is the maximum cut component corresponding to the jth neutral atom, X j The value is 0 or 1, indicating that the jth neutral atom is in component 0 or 1, X k is the maximum cut component corresponding to the kth neutral atom, X k is 0 or 1, indicating that the jth neutral atom is in component 0 or 1, and E is the number of neutral atoms.

[0133] The quantum bit processing module 20 adjusts the laser intensity of the optical tweezers array and the atomic distance between neutral atoms in the optical tweezers array to obtain a target Hamiltonian that meets preset conditions.

[0134] For example, the qubit processing module 20 can respond to user instructions and adjust the laser intensity of the optical tweezers array and the atomic distance between neutral atoms in the optical tweezers array to make Ω, V jk and Δ j By satisfying specific conditions and achieving the desired target Hamiltonian, the solution to the Max-Cut problem is given. For example, by modulating the laser intensity, the equivalent Rabi drive frequency Ω can be achieved, and by adjusting the laser resonant frequency, the detuning amount Δ can be adjusted. j By changing the distance between the two neutral atoms through the optical tweezers array, the interaction strength V between the two neutral atoms can be adjusted. jk , thereby changing the coupling strength of the two quantum bits.

[0135] If Ω, V jk and Δ j The following conditions are met:

[0136] Ω=0;

[0137] V jk =4ω jk ;

[0138]

[0139] Substituting the above parameters into the Hamiltonian, the target Hamiltonian is:

[0140]

[0141] set up:

[0142]

[0143] Then the target Hamiltonian is obtained as:

[0144]

[0145] The qubit processing module 20 solves the maximized loss function based on the target Hamiltonian to obtain two sets of corresponding solutions.

[0146] Optionally, the qubit processing module 20 measures the ground state of the simulated quantum computation based on the target Hamiltonian to obtain two sets of corresponding solutions.

[0147] For example, suppose:

[0148]

[0149] Among them, Z j and Z k The value of is only ±1.

[0150] Substituting the above parameters into the maximum loss function, we get:

[0151] C Max-Cut =∑ j ∑ k ω jk (1-Z j Z k ).

[0152] The qubit processing module 20 solves the maximum loss function based on the above parameters. jk If the individual sums are a constant, then maximizing the loss function requires minimizing the second summation term, i.e., the solution is consistent with the result corresponding to the ground state of the target Hamiltonian. That is, under the above parameters, the qubit processing module 20 can simulate the ground state of quantum computation based on the target Hamiltonian measurement, i.e., can obtain two corresponding sets of solutions that maximize the loss function.

[0153] The qubit processing module 20 performs segmentation processing on the two-qubit gate frequency map based on the two groups of corresponding solutions to obtain two qubit groups.

[0154] For example, the qubit processing module 20 can divide the neutral atoms into two groups of components with the maximum number of connections based on the two corresponding solutions of the maximized loss function, thereby obtaining a cutting diagram of the two-bit gate frequency diagram.

[0155] The quantum circuit compilation module 30 executes two-bit gates between and within the two groups of quantum bits based on the optical tweezers array until all quantum bits are executed.

[0156] For example, the quantum circuit compilation module 30 can implement a two-bit gate between quantum bits based on the coherent movement control of neutral atoms by an optical tweezers array.

[0157] Exemplarily, the quantum circuit compilation module 30 may first execute a two-bit gate of qubits between two qubit groups based on the optical tweezers array, and then execute a two-bit gate of qubits within a qubit group based on the optical tweezers array.

[0158] The quantum circuit compilation module 30 moves the neutral atoms set in the AOD optical tweezers array that are mapped to the first group of quantum bits to the positions of the neutral atoms set in the SLM optical tweezers array that are mapped to the second group of quantum bits, so that a two-bit gate is executed between the first group of quantum bits and the second group of quantum bits.

[0159] For example, when the quantum circuit compilation module 30 executes a two-qubit gate between groups, the qubits in the first qubit group are mapped to neutral atoms arranged in the SLM optical tweezers array, and the qubits in the second qubit group are mapped to neutral atoms arranged in the AOD optical tweezers array. The quantum circuit compilation module 30 can control the movement of the neutral atoms in the AOD optical tweezers array through the AOD optical tweezers array, thereby moving the neutral atoms in the AOD optical tweezers array to the neutral atoms in the SLM optical tweezers array. This configuration allows the two-qubit gate between the two qubit groups to be executed in a maximally parallel manner.

[0160] Optionally, the quantum circuit compilation module 30 determines a target neutral atom mapped to the target qubits in the first and second qubit groups and disposed in the SLM optical tweezers array.

[0161] For example, after executing a two-bit gate between the first qubit group and the second qubit group, all neutral atoms can be set in the SLM optical tweezers array so that the target qubits in the first qubit group and the second qubit group can be mapped to the target neutral atoms of the SLM optical tweezers array.

[0162] The quantum circuit compilation module 30 moves the target neutral atom to a target position in the AOD optical tweezers array based on the AOD optical tweezers array, so that the target quantum bit and the quantum bit mapped by the neutral atom at the target position execute a two-bit gate.

[0163] For example, the quantum circuit compilation module 30 can tweeze a target neutral atom set in the SLM optical tweezers array to the position of another neutral atom each time based on the AOD optical tweezers array, so that the mapped quantum bit can execute a two-bit gate.

[0164] After executing the two-bit gate, the quantum circuit compilation module 30 moves the target neutral atom back to its initial position based on the AOD optical tweezers array.

[0165] The quantum circuit compilation module 30 traverses all target neutral atoms mapped by the target qubits in the first and second qubit groups and set in the SLM optical tweezers array until the two-bit gates of all target qubits are executed.

[0166] For example, after executing the two-bit gate for the target qubit, the quantum circuit compilation module 30 uses the AOD optical tweezers array to move the corresponding neutral atom to its initial position in the SLM optical tweezers array. This process is repeated until the two-bit gates for all qubits have been executed, completing the entire quantum circuit and implementing neutral atom quantum circuit compilation.

[0167] The present invention maps the qubits of a preset quantum circuit to neutral atoms in an optical tweezers array to obtain a two-qubit gate quantum circuit, and constructs a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit. The present invention can determine the Hamiltonian for simulated quantum computing, encode the two-qubit gate frequency map based on a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-qubit gate frequency map, and obtain a target Hamiltonian that meets preset conditions by adjusting the laser intensity of the optical tweezers array and the atomic distances between neutral atoms in the optical tweezers array. The maximized loss function is then solved based on the target Hamiltonian to obtain two corresponding solutions. The two-qubit gate frequency map is segmented based on the two corresponding solutions to obtain two qubit groups. Finally, two-qubit gates can be executed on the two qubit groups, both within and between the groups, using the optical tweezers array, until all qubits are executed.

[0168] Through the above-described embodiments, the present invention utilizes the characteristics of both simulated and digital quantum computing in a neutral atom quantum computing system, enabling the entire quantum circuit compilation process to be completed using only quantum computing power. By simulating quantum computing's ability to solve the maximum cut problem algorithm, the present invention can accelerate the computational speed of the circuit compilation process.

[0169] The present invention can combine analog quantum computing and digital quantum computing to compile neutral atom quantum circuits. No classical computing power is required to implement the entire quantum algorithm. Based on analog quantum computing, some NP problems (such as the maximum cut problem) can be quickly solved. The quantum algorithm is then compiled through digital quantum computing, and the compilation process can be mapped to the maximum cut problem, allowing the solution of the mapping scheme to be completed using analog quantum computing. Furthermore, through the routing design of neutral atoms, the present invention can enable the mapping scheme obtained from analog quantum computing to be run on the digital quantum computing platform, thereby completing the rapid compilation of quantum circuits.

[0170] The present invention can implement the neutral atom quantum circuit compilation method described above by only using one set of SLM equipment and one set of AOD equipment, does not require additional physical equipment restrictions, and has the characteristic of simple structure.

[0171] According to yet another aspect of the present invention, an electronic device is provided. The electronic device includes: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method described above.

[0172] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method described above can be implemented.

[0173] According to another aspect of the present invention, a computer program product is provided, which includes a computer program stored on a computer-readable storage medium; the computer program includes program instructions, which, when executed by a computer, cause the computer to execute the method described above.

[0174] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions of the aforementioned embodiments or substitute equivalents for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A neutral atom quantum circuit compilation method, characterized in that: include: Mapping the qubits of a pre-set quantum circuit to neutral atoms in an optical tweezers array to obtain a two-qubit gate quantum circuit; constructing a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit; Determine the Hamiltonian for simulating quantum computations; Encoding the two-bit gate frequency map based on a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-bit gate frequency map; Adjusting the laser intensity of the optical tweezers array and the atomic distance between neutral atoms in the optical tweezers array to obtain a target Hamiltonian that meets preset conditions; Solving the maximized loss function based on the target Hamiltonian to obtain two sets of corresponding solutions; Segmenting the two-qubit gate frequency map based on the two groups of corresponding solutions to obtain two qubit groups; Based on the optical tweezers array, two-bit gates are executed between and within the two groups of quantum bits respectively until all the quantum bits are executed.

2. The neutral atom quantum circuit compilation method according to claim 1, characterized in that: Determining the Hamiltonian for simulating quantum computing includes: The Hamiltonian of the simulated quantum calculation is determined based on the ground state or Rydberg state of the neutral atoms and the interaction strength between the neutral atoms.

3. The neutral atom quantum circuit compilation method according to claim 1, characterized in that: The encoding of the two-bit gate frequency map based on the maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-bit gate frequency map includes: The maximization loss function is determined according to the maximum cut problem algorithm based on the number of two-bit gate operations between the neutral atoms.

4. The neutral atom quantum circuit compilation method according to claim 1, characterized in that: Solving the maximized loss function based on the target Hamiltonian to obtain two sets of corresponding solutions includes: Based on the target Hamiltonian, the ground state of the simulated quantum computation is measured to obtain the two sets of corresponding solutions.

5. The neutral atom quantum circuit compilation method according to claim 1, characterized in that: The optical tweezers array includes an SLM optical tweezers array and an AOD optical tweezers array, and executing a two-bit gate between and within the two groups of quantum bits based on the optical tweezers array until all the quantum bits are executed includes: Executing the two-bit gate between the groups includes: The neutral atoms set in the AOD optical tweezers array and mapped to the first group of quantum bit groups are moved to the position of the neutral atoms set in the SLM optical tweezers array and mapped to the second group of quantum bit groups, so that the two-bit gate is executed between the first group of quantum bit groups and the second group of quantum bit groups.

6. The neutral atom quantum circuit compilation method according to claim 5, characterized in that: The step of executing two-bit gates between and within the two groups of quantum bits based on the optical tweezers array until all the quantum bits are executed further includes: Executing the two-bit gate within the group comprises: Determining target neutral atoms mapped to target qubits in the first group of qubits and the second group of qubits and disposed in the SLM optical tweezers array; Moving the target neutral atom to a target position in the AOD optical tweezers array based on the AOD optical tweezers array, so that the target quantum bit and the quantum bit mapped by the neutral atom at the target position perform a two-bit gate; After executing the two-bit gate, moving the target neutral atom back to an initial position based on the AOD optical tweezers array; All target neutral atoms mapped by the target qubits in the first group of qubits and the second group of qubits and arranged in the SLM optical tweezers array are traversed until the two-bit gates of all target qubits are executed.

7. A neutral atom quantum circuit compilation system, characterized in that: The neutral atom quantum circuit compilation system is used to execute the method according to any one of claims 1 to 6, and the neutral atom quantum circuit compilation system includes: a quantum circuit processing module that maps the qubits of a preset quantum circuit to neutral atoms in an optical tweezers array to obtain a two-qubit gate quantum circuit, and constructs a two-qubit gate frequency map based on the two-qubit gate quantum circuit in the two-qubit gate quantum circuit; a qubit processing module, which determines a Hamiltonian for simulating quantum computing, encodes the two-bit gate frequency map based on a maximum cut problem algorithm to obtain a maximized loss function corresponding to the two-bit gate frequency map, adjusts the laser intensity of the optical tweezers array and the atomic distance between neutral atoms in the optical tweezers array to obtain a target Hamiltonian that meets preset conditions, solves the maximized loss function based on the target Hamiltonian to obtain two groups of corresponding solutions, and segments the two-bit gate frequency map based on the two groups of corresponding solutions to obtain two qubit groups; The quantum circuit compilation module executes two-bit gates between and within the two groups of quantum bits based on the optical tweezers array until all the quantum bits are executed.

8. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The method comprises a computer program stored on a computer-readable storage medium, wherein the computer program comprises program instructions, and when the program instructions are executed by a computer, the computer is caused to execute the method according to any one of claims 1 to 6.

Images