# US 2026/0099754 A1 · Neural Network-Based Quantum Error Correction Decoding Method and Apparatus, Device, and Chip

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0099754 A1
- Document ID: US20260099754A1
- Status: Published application
- Kind: A1
- Application number: 18/925,709
- Filed: 2024-10-24
- Published: 2026-04-09
- Earliest claimed: 2022-11-22
- Assignees: Tencent Technology (Shenzhen) Company Limited
- Applicants: Tencent Technology (Shenzhen) Company Limited
- Inventors: Yicong ZHENG
- Web page: https://patentgrep.com/patent/US20260099754A1

## CPC

- G06N 10/70: Quantum error correction, detection or prevention, e.g. surface codes or magic state distillation
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 10/20: Models of quantum computing, e.g. quantum circuits or universal quantum computers
- G06N 10/60: Quantum algorithms, e.g. based on quantum optimisation, quantum Fourier or Hadamard transforms
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present disclosure describes neural network-based quantum error correction decoding methods and apparatus, a device, and a chip, relating to the field of artificial intelligence and quantum technologies. One method includes: acquiring error syndrome information obtained from syndrome measurement performed on a quantum circuit; extracting feature information from the error syndrome information by using a neural network decoder; decoding the feature information to obtain a decoding result by using the neural network decoder; and determining error result information of the quantum circuit based on the decoding result.

## Claims

### Claim 1

A method for decoding quantum error correction based on a neural network, performed by an electronic device, the method comprising:
acquiring error syndrome information obtained from syndrome measurement performed on a quantum circuit;
extracting feature information from the error syndrome information by using a neural network decoder;
decoding the feature information to obtain a decoding result by using the neural network decoder; and
determining error result information of the quantum circuit based on the decoding result.

### Claim 2 (depends on 1)

The method according to claim 1, wherein:
the extracting the feature information from the error syndrome information comprises:
using a feature extraction network of the neural network decoder to perform feature extraction on the error syndrome information to obtain the feature information, the neural network decoder comprising the feature extraction network and n feature decoding networks, and n being an integer greater than 1;
the decoding the feature information to obtain the decoding result comprises:
using the n feature decoding networks of the neural network decoder to separately decode the feature information to obtain decoding results respectively corresponding to the n feature decoding networks, the n feature decoding networks being trained in a multi-task learning manner to be enabled to generate different decoding results; and
the determining the error result information of the quantum circuit based on the decoding result comprises:
determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks.

### Claim 3 (depends on 2)

The method according to claim 2, wherein:
a plurality of qubits comprised in the quantum circuit are divided into n blocks, and each block comprises at least one qubit;
for a kth feature decoding network of the n feature decoding networks, a decoding result corresponding to the kth feature decoding network comprises: a Pauli operator acting on a qubit comprised in a kth block of the n blocks, wherein k is a positive integer less than or equal to n; and
the determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks comprises:
determining the error result information based on Pauli operators acting on the n blocks, respectively.

### Claim 4 (depends on 3)

The method according to claim 3, wherein the error result information indicates a qubit in which a Pauli X error occurs and a qubit in which a Pauli Z error occurs in the quantum circuit.

### Claim 5 (depends on 2)

The method according to claim 2, wherein the using the n feature decoding networks of the neural network decoder to separately decode the feature information to obtain the decoding results respectively corresponding to the n feature decoding networks comprises:
using n1 feature decoding networks within the n feature decoding networks to separately decode the feature information to obtain decoding results respectively corresponding to the n1 feature decoding networks, for an ith feature decoding network of the n1 feature decoding networks, a decoding result corresponding to the ith feature decoding network comprising: an ith canonical syndrome related to a target error type, i being a positive integer less than or equal to n1, and the canonical syndrome being a canonical decomposition result of the error syndrome information; and
using n2 feature decoding networks within the n feature decoding networks to separately decode the feature information to obtain decoding results respectively corresponding to the n2 feature decoding networks, for a jth feature decoding network of the n2 feature decoding networks, a decoding result corresponding to the jth feature decoding network comprising: a fixed representative element related to the target error type, j being a positive integer less than or equal to n2,
wherein a sum of n1 and n2 is equal to n, and both n1 and n2 are positive integers.

### Claim 6 (depends on 5)

The method according to claim 5, wherein:
the target error type comprises a Pauli X error and a Pauli Z error;
a sum of m1 and m2 is equal to n1, and both m1 and m2 are positive integers;
m1 feature decoding networks of the n1 feature decoding networks are configured to separately decode the feature information to obtain m1 canonical syndromes related to the Pauli X error;
m2 feature decoding networks of the n1 feature decoding networks are configured to separately decode the feature information to obtain m2 canonical syndromes related to the Pauli Z error;
n2 is equal to 2;
one of the n2 feature decoding networks is configured to decode the feature information to obtain a fixed representative element related to the Pauli X error;
another one of the n2 feature decoding networks is configured to decode the feature information to obtain a fixed representative element related to the Pauli Z error; and
the determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks comprises:
determining X-type error result information based on the fixed representative element related to the Pauli X error and the m1 canonical syndromes related to the Pauli X error, the X-type error result information indicating a qubit in which the Pauli X error occurs in the quantum circuit; and
determining Z-type error result information based on the fixed representative element related to the Pauli Z error and the m2 canonical syndromes related to the Pauli Z error, the Z-type error result information indicating a qubit in which the Pauli Z error occurs in the quantum circuit.

### Claim 7 (depends on 2)

The method according to claim 2, wherein:
the feature extraction network of the neural network decoder comprises a plurality of cascaded feature extraction subnetworks, wherein input data of a first feature extraction subnetwork comprises the error syndrome information, input data of an sth feature extraction subnetwork comprises output data of an (s−1)th feature extraction subnetwork, output data of a last feature extraction subnetwork comprises the feature information, and s is an integer greater than 1;
for a target feature extraction subnetwork of the plurality of cascaded feature extraction subnetworks, input data of the target feature extraction subnetwork is divided into a plurality of input data blocks of a same scale;
the target feature extraction subnetwork is configured to perform a plurality of local feature extraction mappings on the plurality of input data blocks to obtain a plurality of sets of mapping output data, wherein each local feature extraction mapping is configured to perform mapping on regions at a same location in the plurality of input data blocks to obtain a set of mapping output data; and the plurality of local feature extraction mappings are configured to perform mapping on regions at different locations in the plurality of input data blocks to obtain the plurality of sets of mapping output data; and
the target feature extraction subnetwork is further configured to obtain output data of the target feature extraction subnetwork based on the plurality of sets of mapping output data.

### Claim 8 (depends on 2)

The method according to claim 2, wherein:
the feature extraction network and the n feature decoding networks comprised in the neural network decoder are deployed on a same chip.

### Claim 9 (depends on 1)

The method according to claim 1, further comprising:
training the neural network decoder by:
acquiring sample error syndrome information and sample error result information corresponding to the sample error syndrome information;
using a to-be-trained neural network decoder to obtain, based on the sample error syndrome information, predicted decoding results respectively corresponding to the n feature decoding networks;
determining, based on the predicted decoding results respectively corresponding to the n feature decoding networks and label decoding results that respectively correspond to the n feature decoding networks and that is determined based on the sample error result information, loss function values respectively corresponding to the n feature decoding networks;
determining a total loss function value based on the loss function values respectively corresponding to the n feature decoding networks; and
adjusting a parameter of the to-be-trained neural network decoder based on the total loss function value to obtain the trained neural network decoder.

### Claim 10

An apparatus for decoding quantum error correction based on a neural network, the apparatus comprising:
a memory storing instructions; and
a processor in communication with the memory, wherein, when the processor executes the instructions, the processor is configured to cause the apparatus to perform:
acquiring error syndrome information obtained from syndrome measurement performed on a quantum circuit;
extracting feature information from the error syndrome information by using a neural network decoder;
decoding the feature information to obtain a decoding result by using the neural network decoder; and
determining error result information of the quantum circuit based on the decoding result.

### Claim 11 (depends on 10)

The apparatus according to claim 10, wherein:
when the processor is configured to cause the apparatus to perform extracting the feature information from the error syndrome information, the processor is configured to cause the apparatus to perform:
using a feature extraction network of the neural network decoder to perform feature extraction on the error syndrome information to obtain the feature information, the neural network decoder comprising the feature extraction network and n feature decoding networks, and n being an integer greater than 1;
when the processor is configured to cause the apparatus to perform decoding the feature information to obtain the decoding result, the processor is configured to cause the apparatus to perform:
using the n feature decoding networks of the neural network decoder to separately decode the feature information to obtain decoding results respectively corresponding to the n feature decoding networks, the n feature decoding networks being trained in a multi-task learning manner to be enabled to generate different decoding results; and
when the processor is configured to cause the apparatus to perform determining the error result information of the quantum circuit based on the decoding result, the processor is configured to cause the apparatus to perform:
determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks.

### Claim 12 (depends on 11)

The apparatus according to claim 11, wherein:
a plurality of qubits comprised in the quantum circuit are divided into n blocks, and each block comprises at least one qubit;
for a kth feature decoding network of the n feature decoding networks, a decoding result corresponding to the kth feature decoding network comprises: a Pauli operator acting on a qubit comprised in a kth block of the n blocks, wherein k is a positive integer less than or equal to n; and
when the processor is configured to cause the apparatus to perform determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks, the processor is configured to cause the apparatus to perform:
determining the error result information based on Pauli operators acting on the n blocks, respectively.

### Claim 13 (depends on 12)

The apparatus according to claim 12, wherein:
the error result information indicates a qubit in which a Pauli X error occurs and a qubit in which a Pauli Z error occurs in the quantum circuit.

### Claim 14 (depends on 11)

The apparatus according to claim 11, wherein, when the processor is configured to cause the apparatus to perform using the n feature decoding networks of the neural network decoder to separately decode the feature information to obtain the decoding results respectively corresponding to the n feature decoding networks, the processor is configured to cause the apparatus to perform:
using n1 feature decoding networks within the n feature decoding networks to separately decode the feature information to obtain decoding results respectively corresponding to the n1 feature decoding networks, for an ith feature decoding network of the n1 feature decoding networks, a decoding result corresponding to the ith feature decoding network comprising: an ith canonical syndrome related to a target error type, i being a positive integer less than or equal to n1, and the canonical syndrome being a canonical decomposition result of the error syndrome information; and
using n2 feature decoding networks within the n feature decoding networks to separately decode the feature information to obtain decoding results respectively corresponding to the n2 feature decoding networks, for a jth feature decoding network of the n2 feature decoding networks, a decoding result corresponding to the jth feature decoding network comprising: a fixed representative element related to the target error type, j being a positive integer less than or equal to n2,
wherein a sum of n1 and n2 is equal to n, and both n1 and n2 are positive integers.

### Claim 15 (depends on 14)

The apparatus according to claim 14, wherein:
the target error type comprises a Pauli X error and a Pauli Z error;
a sum of m1 and m2 is equal to n1, and both m1 and m2 are positive integers;
m1 feature decoding networks of the n1 feature decoding networks are configured to separately decode the feature information to obtain m1 canonical syndromes related to the Pauli X error;
m2 feature decoding networks of the n1 feature decoding networks are configured to separately decode the feature information to obtain m2 canonical syndromes related to the Pauli Z error;
n2 is equal to 2;
one of the n2 feature decoding networks is configured to decode the feature information to obtain a fixed representative element related to the Pauli X error;
another one of the n2 feature decoding networks is configured to decode the feature information to obtain a fixed representative element related to the Pauli Z error; and
when the processor is configured to cause the apparatus to perform determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks, the processor is configured to cause the apparatus to perform:
determining X-type error result information based on the fixed representative element related to the Pauli X error and the m1 canonical syndromes related to the Pauli X error, the X-type error result information indicating a qubit in which the Pauli X error occurs in the quantum circuit; and
determining Z-type error result information based on the fixed representative element related to the Pauli Z error and the m2 canonical syndromes related to the Pauli Z error, the Z-type error result information indicating a qubit in which the Pauli Z error occurs in the quantum circuit.

### Claim 16 (depends on 11)

The apparatus according to claim 11, wherein:
the feature extraction network of the neural network decoder comprises a plurality of cascaded feature extraction subnetworks, wherein input data of a first feature extraction subnetwork comprises the error syndrome information, input data of an sth feature extraction subnetwork comprises output data of an (s−1)th feature extraction subnetwork, output data of a last feature extraction subnetwork comprises the feature information, and s is an integer greater than 1;
for a target feature extraction subnetwork of the plurality of cascaded feature extraction subnetworks, input data of the target feature extraction subnetwork is divided into a plurality of input data blocks of a same scale;
the target feature extraction subnetwork is configured to perform a plurality of local feature extraction mappings on the plurality of input data blocks to obtain a plurality of sets of mapping output data, wherein each local feature extraction mapping is configured to perform mapping on regions at a same location in the plurality of input data blocks to obtain a set of mapping output data; and the plurality of local feature extraction mappings are configured to perform mapping on regions at different locations in the plurality of input data blocks to obtain the plurality of sets of mapping output data; and
the target feature extraction subnetwork is further configured to obtain output data of the target feature extraction subnetwork based on the plurality of sets of mapping output data.

### Claim 17 (depends on 11)

The apparatus according to claim 11, wherein:
the feature extraction network and the n feature decoding networks comprised in the neural network decoder are deployed on a same chip.

### Claim 18 (depends on 10)

The apparatus according to claim 10, wherein, when the processor executes the instructions, the processor is further configured to cause the apparatus to perform:
training the neural network decoder by:
acquiring sample error syndrome information and sample error result information corresponding to the sample error syndrome information;
using a to-be-trained neural network decoder to obtain, based on the sample error syndrome information, predicted decoding results respectively corresponding to the n feature decoding networks;
determining, based on the predicted decoding results respectively corresponding to the n feature decoding networks and label decoding results that respectively correspond to the n feature decoding networks and that is determined based on the sample error result information, loss function values respectively corresponding to the n feature decoding networks;
determining a total loss function value based on the loss function values respectively corresponding to the n feature decoding networks; and
adjusting a parameter of the to-be-trained neural network decoder based on the total loss function value to obtain the trained neural network decoder.

### Claim 19

A non-transitory computer-readable storage medium, storing computer-readable instructions, wherein, the computer-readable instructions, when executed by a processor, are configured to cause the processor to perform:
acquiring error syndrome information obtained from syndrome measurement performed on a quantum circuit;
extracting feature information from the error syndrome information by using a neural network decoder;
decoding the feature information to obtain a decoding result by using the neural network decoder; and
determining error result information of the quantum circuit based on the decoding result.

### Claim 20 (depends on 19)

The non-transitory computer-readable storage medium according to claim 19, wherein:
when the computer-readable instructions are configured to cause the processor to perform extracting the feature information from the error syndrome information, the computer-readable instructions are configured to cause the processor to perform:
using a feature extraction network of the neural network decoder to perform feature extraction on the error syndrome information to obtain the feature information, the neural network decoder comprising the feature extraction network and n feature decoding networks, and n being an integer greater than 1;
when the computer-readable instructions are configured to cause the processor to perform decoding the feature information to obtain the decoding result, the computer-readable instructions are configured to cause the processor to perform:
using the n feature decoding networks of the neural network decoder to separately decode the feature information to obtain decoding results respectively corresponding to the n feature decoding networks, the n feature decoding networks being trained in a multi-task learning manner to be enabled to generate different decoding results; and
when the computer-readable instructions are configured to cause the processor to perform determining the error result information of the quantum circuit based on the decoding result, the computer-readable instructions are configured to cause the processor to perform:
determining the error result information based on the decoding results respectively corresponding to the n feature decoding networks.
