IP Library › Granted Patent US 12,079,691
Granted Patent B2
US 12,079,691 · App. 18/278,723 · Granted Sep 3, 2024

Quantum convolution operator

Inventors: Menghan Dou (Anhui, CN); Yuan Fang (Anhui, CN); Zhaohui Zhou (Anhui, CN); Hanchao Wang (Anhui, CN); Lei Li (Anhui, CN)
Assignee: Origin Quantum Computing Technology (Hefei) Co., Ltd.
G06N10/20
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,079,691
App. No.
18/278,723
Granted
Sep 3, 2024
Kind
B2
Abstract

The present disclosure provides a quantum convolution operator, comprising: a quantum state encoding module, a quantum entanglement module, a quantum convolution kernel module, a measuring module, and a computing module; the quantum state encoding module is configured to encode a current group of input data onto qubits; the quantum entanglement module is configured to associate quantum state information of different qubits; the quantum convolution kernel module is configured to extract feature information corresponding to the quantum state information; the measuring module is configured to measure a quantum state of a preset qubit and obtain a corresponding amplitude; the computing module is configured to compute a convolution result corresponding to the current group of input data according to the measured quantum state and its amplitude.

Claims (80)

1. A quantum computer, comprising:

a processor; and

a memory having computer program stored thereon, wherein the computer program, when being run by the processor, implements a quantum convolution operator; wherein

the quantum convolution operator comprises: a quantum state encoding module, a quantum entanglement module, a quantum convolution kernel module, a measuring module, and a computing module; wherein

the quantum state encoding module is configured to encode a current group of input data onto qubits;

the quantum entanglement module comprises CNOT gates acting on two adjacent qubits, and is configured to associate quantum state information of different qubits via the CNOT gates;

the quantum convolution kernel module is configured to extract feature information corresponding to the quantum state information;

the measuring module is configured to measure a quantum state of a preset qubit and obtain a corresponding amplitude; and

the computing module is configured to compute a convolution result corresponding to the current group of input data according to the measured quantum state and its amplitude;

wherein the quantum convolution kernel module comprises a U3 gate of a quantum logic gate, wherein a unitary matrix of the U3 gate is:

R

⁡

(

α

,

β

,

γ

)

=

[

cos

⁡

(

α

2

)

-

e

i

⁢

γ

×

sin

⁡

(

α

2

)

e

i

⁢

β

×

sin

⁡

(

α

2

)

e

i

⁢

γ

+

i

⁢

β

×

cos

⁡

(

α

2

)

]

wherein α, β, and γ are angular variables.

2. The quantum computer of claim 1 , wherein the quantum state encoding module comprises quantum logic gates; and

said “encode the current group of input data onto qubits”, comprises:

determining a variable parameter value in a quantum logic gate corresponding to the current group of input data, wherein the quantum logic gate supports dynamic modification of the variable parameter value.

3. The quantum computer of claim 1 , wherein the quantum convolution operator further comprises:

an updating module, configured to update a parameter value of a quantum state encoding module corresponding to the next group of input data according to a preset mapping relationship.

4. The quantum computer of claim 3 , wherein said “update a parameter value of a quantum state encoding module corresponding to the next group of input data according to a preset mapping relationship”, comprises:

according to a preset mapping relationship between the input data and quantum state encoding circuits of the quantum state encoding module, updating a quantum state encoding circuit corresponding to the next group of input data so as to obtain a new quantum state encoding circuit as an updated quantum state encoding module.

5. The quantum computer of claim 1 , wherein said “compute a convolution result corresponding to the input data according to the measured quantum state and its amplitude”, comprises:

according to the measured quantum state and its amplitude, computing a probability corresponding to when the preset qubit is in a preset quantum state as a convolution result corresponding to the current group of input data.

6. The quantum computer of claim 1 , wherein in the case where the quantum entanglement module and/or the quantum convolution kernel module comprises a quantum logic gate, a variable parameter value in the quantum logic gate is determined by pre-training.

7. The quantum computer of claim 1 , wherein the input data comprises: an image, natural language, or audio data.

8. The quantum computer of claim 1 , wherein the quantum state comprises a following quantum logic gate: an RX gate, an RY gate, an RZ gate, or an X gate.

9. The quantum computer of claim 1 , wherein the last CNOT gate acts on the last qubit and the first qubit.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 24, 2023
From: DOU, MENGHAN; FANG, YUAN; ZHOU, ZHAOHUI; WANG, HANCHAO; LI, LEI
To: ORIGIN QUANTUM COMPUTING TECHNOLOGY (HEFEI) CO., LTD.
Reel/Frame 064694/0153 →
Priority Claims (6)
CN 202110213240.X · Feb 26, 2021 · national
CN 202110339877.3 · Mar 30, 2021 · national
CN 202110473480.3 · Apr 29, 2021 · national
CN 202110475104.8 · Apr 29, 2021 · national
CN 202110722243.6 · Jun 28, 2021 · national
CN 202111683374.4 · Dec 31, 2021 · national
Continuity (1)
Related Publication 20240095563A1 · Mar 21, 2024
Cited By (1)
US 12,596,951