Training method and device for generative adversarial network model, equipment, program and storage medium

Patent metadata

Field	Value
Publication number	US-12456056-B2
Application number	US-202117457903-A
Country	US
Kind code	B2
Filing date	Dec 6, 2021
Priority date	Dec 24, 2020
Publication date	Oct 28, 2025
Grant date	Oct 28, 2025

Abstract

Provided are a training method and device for a heterogeneous generative adversarial network model, an equipment, a program and a storage medium. In the training method, measurement data of a heterogeneous station is acquired, the measurement data of the heterogeneous station is set as a training sample, and joint training is performed on the heterogeneous generative adversarial network model according to a total objective function. A generator is configured to predict environment data at a future occasion according to environment data of the heterogeneous station at a historical occasion so as to output predicted data. A discriminator is configured to be input the predicted data output by the generator and corresponding measurement data, and discriminate a similarity between the measurement data and the predicted data; a total objective function includes a first objective function of the generator and a second objective function of the discriminator.

First claim

What is claimed is:

1. A training method for a heterogeneous generative adversarial network model, executed by electronic equipment, wherein the heterogeneous generative adversarial network model comprises a generator and a discriminator, and the method comprises:
acquiring measurement data of a heterogeneous station, wherein the heterogeneous station comprises at least two types of stations, and each type of station among the at least two types of stations is configured to measure and obtain environment data corresponding to the each type as measurement data of the each type of station; and
setting the measurement data of the heterogeneous station as a training sample, and performing joint training on the heterogeneous generative adversarial network model according to a total objective function;
wherein the heterogeneous station comprises an air quality monitoring station and a weather monitoring station, and the measurement data comprises weather measurement data and air quality measurement data;
wherein the generator is configured to predict environment data at a future occasion according to environment data of the heterogeneous station at a historical occasion so as to output predicted data;
the discriminator is configured to be input the predicted data output by the generator and corresponding measurement data, and discriminate a similarity between the measurement data and the predicted data; and
the total objective function comprises a first objective function of the generator and a second objective function of the discriminator;
wherein the discriminator comprises at least one of:
a space discriminator, which is configured to be input predicted data output by the generator and corresponding measurement data of each station at a set occasion, and discriminate a space similarity between the measurement data and the predicted data at the set occasion;
a time discriminator, which is configured to be input predicted data output by the generator and corresponding measurement data of a set station at at least two occasions, and discriminate a time similarity between the measurement data and the predicted data of the set station; or
a macro discriminator, which is configured to be input predicted data output by the generator and corresponding measurement data of a plurality of stations at multiple occasions, and discriminate a macro similarity between the predicted data and the measurement data of the generator;
wherein the discriminator is provided with the second objective function.

2. The method according to claim 1 , wherein the total objective function is a weighted sum of second objective functions of a plurality of discriminators plus an accumulated sum of the first objective function.

3. The method according to claim 2 , wherein a formula of the total objective function is expressed as follows:
L=L g +Σ i=1 K λ i L d i ,

wherein L is the total objective function, L g is the first objective function of the generator, L d i is a second objective function of a discriminator d i , λ i is a weight of the discriminator d i , the discriminator d i is the space discriminator, the time discriminator or the macro discriminator, and K is a number of discriminators.

4. The method according to claim 1 , wherein the space discriminator comprises a graph neural network, a pooling layer and a multilayer perceptron;
input data of the space discriminator is a prediction matrix formed by predicted data of each heterogeneous station at the set occasion and a measurement matrix formed by measurement data of the each heterogeneous station at the set occasion, and the multilayer perceptron is configured to calculate a probability that the input data is the measurement data; and
a formula of a second objective function L s of the space discriminator is expressed as follows:
L s =log D s (Y t ;θ s )+log(1−D s (Ŷ t ;θ s )),

wherein D s ( ) is a space discriminator function, Y t is a measurement matrix, Ŷ t is a prediction matrix, and θ s is a to-be-trained parameter of the space discriminator.

5. The method according to claim 1 , wherein the time discriminator comprises a gated recurrent model and a multilayer perceptron;
input data of the time discriminator is a prediction sequence formed by predicted data of a set station at a plurality of occasions and a measurement sequence formed by measurement data of the set station at the plurality of occasions, and the multilayer perceptron is configured to calculate a probability that the input data is the measurement data; and
a formula of a second objective function L t of the time discriminator is as follows:
L t =log D t (y i ;θ t )+log(1−D t (ŷ i ;θ t )),

wherein D t ( ) is a time discriminator function, y; is a measurement sequence, ŷ i is a prediction sequence, and θ t is a to-be-trained parameter of the time discriminator.

6. The method according to claim 1 , wherein the macro discriminator comprises a connecting layer, a gated recurrent model and a multilayer perceptron;
input data of the macro discriminator is a prediction matrix formed by predicted data of each heterogeneous station at a plurality of occasions and a measurement matrix formed by measurement data of the each heterogeneous station at the plurality of occasions;
the connecting layer is configured to splice a vector of each station in each matrix into an occasion vector, and the multilayer perceptron is configured to calculate a probability that the input data is the measurement data; and
a formula of a second objective function L m of the macro discriminator is expressed as follows:
L m =logD m (Y;θ m )+log(1−D m (Ŷ;θ m )),

wherein D m ( ) is a macro discriminator function, Y is a vector of all stations at a prediction occasion, Ŷ is a vector of the all stations at a measurement occasion, and θ m is a to-be-trained parameter of the time discriminator.

7. The method according to claim 3 , wherein a formula for calculating the weight λ i is expressed as follows:

λ
i

=

exp
⁡
(

γ
i

)

∑

k
=
1

K

⁢

exp
⁡
(

γ
k

)

,

a formula for calculating γ i is expressed as follows:
γ i =sim(σ(H i ),σ(H i )),

wherein sim( ) is a function based on an Euclidean distance, σ represents a sigmoid function, H i is a hidden layer vector of the discriminator d i when measurement data is input, and Ĥ i is a hidden layer vector of the discriminator d i when predicted data is input.

8. The method according to claim 1 , wherein the model of the generator is constructed based on a heterogeneous station graph, nodes in the heterogeneous station graph represent stations, types of the stations comprise an air quality monitoring stat

Assignees

• Beijing Baidu Netcom Sci & Tech Co Ltd

Inventors

• Liu Hao
• Han Jindong
• Zhu Hengshu
• Dou Dejing

Classifications

• G06N3/045

  Combinations of networks · CPC title

• G01W1/02 Primary

  Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed (G01W1/10 takes precedence) · CPC title

• G01N33/0009

  General constructional details of gas analysers, e.g. portable test equipment (devices for withdrawing samples in the gaseous state G01N1/22) · CPC title

• G06N3/0475

  Generative networks · CPC title

• G06N3/09

  Supervised learning · CPC title