CN114781496B - Optimizing sampling method and device and electronic equipment - Google Patents
Optimizing sampling method and device and electronic equipment Download PDFInfo
- Publication number
- CN114781496B CN114781496B CN202210349040.1A CN202210349040A CN114781496B CN 114781496 B CN114781496 B CN 114781496B CN 202210349040 A CN202210349040 A CN 202210349040A CN 114781496 B CN114781496 B CN 114781496B
- Authority
- CN
- China
- Prior art keywords
- value
- difference
- characteristic
- pair
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000005070 sampling Methods 0.000 title claims abstract description 34
- 210000002569 neuron Anatomy 0.000 claims description 104
- 239000011159 matrix material Substances 0.000 claims description 34
- 230000006870 function Effects 0.000 claims description 17
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 10
- 238000010801 machine learning Methods 0.000 abstract description 3
- 230000000875 corresponding effect Effects 0.000 description 21
- 238000013528 artificial neural network Methods 0.000 description 16
- 238000004590 computer program Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 7
- 230000004913 activation Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 230000014759 maintenance of location Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000033228 biological regulation Effects 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Image Analysis (AREA)
Abstract
The disclosure provides an optimizing sampling method, an optimizing sampling device and electronic equipment, relates to the technical field of data processing, and particularly relates to the technical field of machine learning. The specific implementation scheme is as follows: acquiring a plurality of objects to be grouped respective object features; for a pair of feature values consisting of feature values of the same feature dimension in every two of the object features, carrying out weighted summation on all the object features to obtain the absolute value of the difference value of the two feature values in the feature value pair; calculating the similarity between every two objects to be grouped according to the absolute value of the difference value; and selecting at least one pair of objects to be grouped, wherein the similarity is larger than a preset similarity threshold value, as an optimized object pair. The efficiency of optimizing and sampling can be improved.
Description
Technical Field
The present disclosure relates to the field of data processing technologies, and in particular, to the field of machine learning technologies.
Background
In some application scenarios, the objects involved in the test, such as the personnel involved in the test, sample data, etc., are divided into experimental groups and control groups. Experiments were performed on the experimental group and the control group, respectively, and variables were introduced in the test to determine the influence of the variables on the experimental results. This experimental method is called the A/B experiment.
Disclosure of Invention
The present disclosure provides a method, apparatus, device, and storage medium for optimizing sampling.
According to an aspect of the present disclosure, there is provided an optimizing sampling method, including:
acquiring respective object characteristics of a plurality of objects to be grouped;
for the characteristic value pairs formed by characteristic values of the same characteristic dimension in every two object characteristics, carrying out weighted summation on all the object characteristics to obtain the difference absolute values of the two characteristic values in the characteristic value pairs;
calculating the similarity between the characteristics of each two objects to be grouped according to the absolute value of the difference;
and selecting at least one pair of objects to be grouped, wherein the similarity is larger than a preset similarity threshold value, as an optimized object pair.
According to a second aspect of the present disclosure, there is provided an optimizing sampling device comprising:
the characteristic acquisition module is used for acquiring the object characteristics of each of a plurality of objects to be grouped;
the difference solving module is used for carrying out weighted summation on all the object features aiming at the feature value pairs formed by the feature values of the same feature dimension in every two object features to obtain the difference absolute values of the two feature values in the feature value pairs;
the similarity solving module is used for calculating the similarity between the characteristics of each two objects to be grouped according to the absolute value of the difference value;
And the object pair screening module is used for selecting at least one pair of objects to be grouped, the similarity of which is greater than a preset similarity threshold value, as an optimizing object pair.
According to a third aspect provided by the present disclosure, there is provided an electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect described above.
According to a fourth aspect of the present disclosure, there is provided a non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method according to the first aspect described above.
According to a fourth aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements a method according to the first aspect described above.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
Drawings
The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
FIG. 1 is a schematic flow chart of the optimizing sampling method provided by the present disclosure;
FIG. 2 is a schematic diagram of a structure of a feed-forward neural network for implementing optimal sampling provided by the present disclosure;
FIG. 3 is a schematic diagram of a difference solving unit in a feedforward neural network according to the present disclosure;
FIG. 4 is another flow diagram of the optimizing sampling method provided by the present disclosure;
FIG. 5 is a schematic view of another configuration of the optimizing sampling device provided by the present disclosure;
fig. 6 is a block diagram of an electronic device for implementing the optimized sampling method of an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
For more clear explanation of the optimizing and sampling method provided by the present disclosure, one possible application scenario of the optimizing and sampling method provided by the present disclosure will be illustrated in the following, and it should be understood that the following examples are only one possible application scenario of the optimizing and sampling method provided by the present disclosure, and in other possible embodiments, the optimizing and sampling method provided by the present disclosure may also be applied to other possible embodiments, and the following examples do not limit any limitation.
An application developer develops a new function for an application, and in order to determine whether to bring the new function on line, the application developer can determine the influence of the new function on the retention rate of a user through an A/B experiment. Illustratively, the plurality of testers are divided into a control group and an experimental group, and the testers of the control group use applications without new functions, and the testers of the experimental group use applications with new functions. And respectively counting the user retention rates of the control group and the experimental group, thereby obtaining the influence of the new function on the user retention rate.
In order to avoid mutual interference between the control group and the experimental group, the same tester cannot belong to both the control group and the experimental group. And in order to improve the accuracy of the experimental results, the variables between the control group and the experimental group should be reduced as much as possible, so that the testers in the control group and the testers in the experimental group need to be similar as much as possible.
It can be seen that if one tester is divided into a control group, another tester sufficiently similar to the tester needs to be divided into an experimental group. Accordingly, in the related art, the testers are often divided into a plurality of tester pairs, each tester pair is composed of two testers (groups) that are sufficiently similar, one of the testers (groups) is assigned to the control group, and the other tester (groups) is assigned to the experiment group, and the process is called optimizing sampling.
In order to accurately divide the testers into a plurality of tester pairs, it is necessary to determine whether each two testers are sufficiently similar, that is, calculate the similarity between each two testers. When the number of testers is large, the similarity required to be calculated is large, so that the efficiency of optimizing and sampling is low.
In the related art, in order to improve the efficiency of optimizing and sampling, only the similarity among part of testers is calculated, and pairs of testers are extracted according to the calculated similarity among part of testers. However, since only some of the testers are calculated for similarity, the testers in the extracted tester pair may not be similar enough, i.e., the accuracy of the optimizing sample is low.
Based on this, the present disclosure provides an optimizing sampling method, as shown in fig. 1, including:
s101, acquiring respective object characteristics of a plurality of objects to be grouped.
S102, for the characteristic value pairs formed by the characteristic values of the same characteristic dimension in every two object characteristics, weighting and summing all the object characteristics to obtain the difference absolute values of the two characteristic values in the characteristic value pairs.
S103, calculating the similarity between the characteristics of each two objects to be grouped according to the absolute value of the difference.
S104, selecting at least one pair of objects to be grouped, the similarity of which is larger than a preset similarity threshold value, as an optimizing object pair.
The embodiment is selected, the calculation of the similarity can be disassembled into the calculation of the absolute value of the difference, and the absolute value of the difference of the two characteristic values in each characteristic value pair is calculated by carrying out weighted summation on all the object characteristics, so that the calculation mode of each absolute value of the difference is the same, and the input is all the object characteristics, and the difference is only that the weights used in the weighted summation process are different when different absolute values of the difference are calculated, so that the absolute values of the difference of the two characteristic values in all the characteristic value pairs can be obtained in batches through one-time input and the same calculation mode, the calculation efficiency of the absolute value of the difference is effectively improved, and the calculation efficiency of the similarity is improved, namely the optimizing sampling efficiency is effectively improved.
On the other hand, the optimizing sampling method provided by the disclosure can calculate the absolute values of the difference values of the two characteristic values in all characteristic value pairs in batches, so that the similarity between every two objects to be grouped can be calculated, and therefore, the objects to be grouped which are similar enough can be accurately determined to be used as optimizing object pairs, and the accuracy of optimizing sampling can be effectively improved.
The aforementioned objects to be grouped may be different types of objects according to application scenes, including but not limited to people, images, texts, etc. And each object to be grouped can be a single individual or a set of a plurality of individuals, for example, each object to be grouped can be a person or a traffic bucket consisting of a plurality of persons.
The aforementioned object features are vectors that characterize the objects to be grouped, each object feature being an M-dimensional vector. In one possible embodiment, in order to enable the feature vector to accurately characterize the objects to be grouped, data generated by each object to be grouped within a preset time length is collected with the objects to be grouped as granularity, and features of the data are extracted as object features of the objects to be grouped.
The size of the similarity herein refers to the size of the degree of similarity represented by the similarity, and not the size in terms of the similarity value. The magnitude of the similarity degree represented by the similarity and the magnitude of the similarity value can be positively correlated or negatively correlated according to different application scenes.
For example, if the cosine distance between the object features of two objects to be grouped is used to represent the similarity between the two objects to be grouped, the larger the value of the similarity is, the larger the degree of similarity represented by the similarity is. If the Euclidean distance is used for representing the similarity between two objects to be grouped, the larger the value of the similarity is, the smaller the similarity degree represented by the similarity is.
It will be appreciated that in the foregoing S102, multiple weighted sums are required, and each weighted sum is performed on all object features, and the operation implemented by the fully-connected layer is performed on each input value multiple weighted sums, so that the foregoing S102 may be implemented using the fully-connected layer.
Illustratively, the foregoing S102 is implemented by:
inputting the characteristic value pairs into a preset full-connection layer aiming at the characteristic value pairs formed by the characteristic values of the same characteristic dimension in each two object characteristics to obtain the absolute value of the difference value of the two characteristic values in the characteristic value pairs output by the full-connection layer,
The full connection layer is used for calculating the absolute value of the difference value of the two characteristic values in the characteristic value pair aiming at each characteristic value pair, and outputting the absolute value of the difference value.
By adopting the embodiment, the absolute value of the difference value can be calculated by using the full-connection layer, on one hand, the full-connection layer is simpler in architecture, and the realization difficulty of the optimizing sampling method provided by the disclosure can be reduced. On the other hand, the architecture of the full connection layer is suitable for parallel computing and GPU computing, so that the efficiency of determining the absolute value of the difference value can be further improved by utilizing modes of parallel computing, GPU computing and the like, namely, the efficiency of the optimizing sampling method provided by the disclosure is further improved.
The full-connection layer can be an independent full-connection layer, or can be a full-connection layer in a preset feedforward neural network, wherein the preset feedforward neural network comprises an input layer, a full-connection layer and an output layer.
For the case that the full-connection layer is in the preset feedforward neural network, the input of the preset feedforward neural network is all object characteristics, and the output can be the absolute value of the difference value of two characteristic values in each characteristic value pair or the similarity between every two objects to be grouped.
In order to more clearly describe the optimizing and sampling method provided in the present disclosure, the structure of the feedforward neural network will be described below by taking the output of the preset feedforward neural network as the similarity between every two objects to be grouped as an example:
Referring to fig. 2, fig. 2 is a schematic structural diagram of a feedforward neural network according to an embodiment of the present invention, where the feedforward neural network includes an input layer, a fully-connected layer, and an output layer. And, the full connection layer includes a plurality of difference solving units.
The input layer is used for inputting the characteristic values of the same characteristic dimension in every two object characteristics as a characteristic value pair to the difference value solving units corresponding to the characteristic value pair, wherein different characteristic value pairs correspond to different difference value solving units.
For example, assume that there are three total objects to be grouped, respectively denoted as objects to be grouped 1-3, where object characteristics of object to be grouped 1 are { x11, x12, x13}, x11 is a characteristic value of object to be grouped 1 in characteristic dimension 1, x12 is a characteristic value of object to be grouped 1 in characteristic dimension 2, and x13 is a characteristic value of object to be grouped 1 in characteristic dimension 3. The object characteristics of the object to be grouped 2 are { x21, x22, x23}, and the object characteristics of the object to be grouped 3 are { x31, x32, x33}.
Then in this embodiment there are a total of 9 eigenvalue pairs, { x11, x21}, { x11, x31}, { x21, x31}, { x12, x22}, { x12, x32}, { x22, x32}, { x13, x23}, { x13, x33}, and { x23, x33}, respectively. Therefore, the input layer inputs the 9 feature values to different difference solving units, respectively.
The difference value solving unit is used for calculating the absolute value of the difference value of the two characteristic values in the characteristic value pair input to the difference value solving unit, and inputting the absolute value of the difference value to the output layer. Illustratively, assuming that { x11, x21} is input to the difference solving unit 1, the difference solving unit 1 calculates the absolute value of the difference between x11 and x21, and inputs the calculated absolute value of the difference to the output layer.
By adopting the embodiment, the full connection layer is unitized, so that the full connection layer is designed according to a specific application scene, and the adaptability of the optimizing sampling method is effectively improved.
The input layers may be one or more layers, and the input of each neuron in the first input layer should be one feature value in one object feature, and the feature values input to different neurons should be different. The last input layer is connected to the fully connected layer, so that each neuron in the fully connected layer is connected to all neurons in the last input layer, respectively.
And the difference solving unit belongs to the full-connection layer, so that each neuron in the last input layer is connected with each difference solving unit. Therefore, by reasonably setting the weight of the feedforward neural network, the output layer can input the characteristic value of the same characteristic dimension in every two object characteristics as a characteristic value pair to the difference value solving unit corresponding to the characteristic value pair.
Exemplary, if the input of each neuron in the last input layer is a characteristic value, the value input from the ith neuron in the input layer to the jth difference solving unit is alpha ij x i Wherein x is i For the eigenvalues of the ith neuron input to the input layer, α ij Weights set for the ith neuron and the jth difference solving unit of the input layer, and if the pair of eigenvalues corresponding to the jth difference solving unit comprises x i Alpha is then ij =1, if the pair of eigenvalues corresponding to the jth difference solving unit does not include x i Alpha is then ij =0. For example, if x i =x12, and the eigenvalue pair corresponding to the jth difference solving unit is { x12, x32}, α ij =1, and if x i =x12, and the eigenvalue pair corresponding to the jth difference solving unit is { x22, x32}, α ij =0. It will be appreciated that if alpha ij =1, the output of the ith neuron is x i Thus, this time, it is considered that the ith neuron will x i Input to the j-th difference solving unit, if alpha ij The output of the ith neuron is 0, so that the ith neuron is regarded as not outputting x i And inputting the difference value to a j-th difference value solving unit.
For the case that the fully-connected layer is an independent fully-connected layer, the fully-connected layer may also include a plurality of difference value solving units, and the principle of the difference value solving units in this case is identical to that of the case that the fully-connected layer is in the preset feedforward neural network, so the difference value solving units in this case may refer to the related description and are not repeated herein.
The structure of the difference solving unit may be different according to the application scenario, and illustratively, in one possible embodiment, the difference solving unit is formed by a neuron, the input of the neuron is two eigenvalues in the eigenvalue pair corresponding to the difference solving unit, the neuron is used for calculating the difference delta of the two eigenvalues, determining a larger value of delta and-delta, and outputting the larger value as the absolute value of the difference of the two eigenvalues. For example, if the characteristic values input to the neuron are 1 and 2, the neuron calculates 1-2= -1, compares the larger values of-1 and 1 to obtain 1, and outputs 1 as the absolute value of the difference between 1 and 2.
In another possible embodiment, as shown in fig. 3, the difference solving unit includes: a first difference neuron, a second difference neuron, and a summing neuron. In this embodiment, the full-connection layer is two layers, and the first difference neurons and the second difference neurons in all the difference solving units form a first full-connection layer, and the summation neurons in all the difference solving units form a second full-connection layer.
In this example, the step of inputting the feature value pair to the preset full connection layer to obtain the absolute value of the difference between the two feature values in the feature value pair output by the full connection layer is implemented in the following manner:
S1021, respectively inputting the characteristic value pairs into a first difference value neuron and a second difference value neuron in a difference value solving unit corresponding to the characteristic value pairs, and obtaining a first output value output by the first difference value neuron and a second output value output by the second difference value neuron.
S1022, inputting the first output value and the second output value into a summation neuron in a difference value solving unit corresponding to the characteristic value pair, and obtaining an output value output by the summation neuron as an absolute value of a difference value of two characteristic values in the characteristic value pair.
In each difference solving unit, the inputs of the first difference neuron and the second difference neuron are as follows: and inputting two eigenvalues in the eigenvalue pair of the difference solving unit.
And the first difference value neuron is used for subtracting one characteristic value from the other characteristic value to obtain a first difference value, judging whether the first difference value is larger than 0, outputting the first difference value if the first difference value is larger than 0, and outputting 0 if the first difference value is not larger than 0.
And the second difference value neuron is used for subtracting the one characteristic value from the other characteristic value to obtain a second difference value, judging whether the second difference value is larger than 0, outputting the second difference value if the second difference value is larger than 0, and outputting 0 if the second difference value is not larger than 0.
And a summation neuron for calculating a summation result of the first output value and the second output value, and outputting the summation result.
For example, assuming that the two eigenvalues are 1,2, respectively, the first difference neuron calculates 1-2, resulting in a first difference of-1, and since-1 is not greater than 0, 0 is output, i.e., the first output value is 0. The second difference neuron calculates 2-1 to obtain a second difference value of 1, and since 1 is greater than 0, 1 is output, i.e., the second output value is 1. The summing neuron calculates 0+1 to obtain a summation result of 1, and outputs the summation result, so that the absolute value of the difference between the two eigenvalues is 1.
It can be understood that the output of different results according to the magnitude relation with the preset threshold is an operation that can be realized by the activation function (herein called an activation function operation), so that the embodiment is selected and used, the calculation of the absolute value of the difference can be realized only by addition (subtraction operation can be regarded as special addition operation) and activation function operation, the characteristics that the full connection layer can effectively realize addition operation and activation function operation are fully utilized, and the efficiency of calculating the absolute value of the difference by the difference solving unit is improved, so that the efficiency of optimizing and sampling is further improved.
The similarity calculated in S103 may be expressed in different forms, such as an array, a matrix, etc., in different application scenarios. For convenience of description, the following will take a matrix as an example:
in one possible embodiment, the foregoing S103 is implemented by:
and generating a similarity matrix according to the absolute value of the difference, wherein the similarity matrix is an N-by-N-dimensional matrix, and N is the number of objects to be grouped. The element of the ith row and the jth column in the similarity matrix is the similarity between the ith object to be grouped and the jth object to be grouped, and i and j are positive integers with the value range of [1, N ]. For example, assuming that there are a total of 4 objects to be grouped, denoted as S1-4, respectively, the similarity matrix is:
<S1，S1> <S1，S2> <S1，S3> <S1，S4>
<S2，S1> <S2，S2> <S2，S3> <S2，S4>
<S3，S1> <S3，S2> <S3，S3> <S3，S4>
<S4，S1> <S4，S2> <S4，S3> <S4，S4>
wherein < S1, S1> is the similarity between S1 and S1, < S1, S2> is the similarity between S1 and S2, and so on.
The embodiment is selected, the similarity between every two objects to be grouped is expressed in a matrix form, and the subsequent processing of the similarity based on matrix batch is facilitated.
It can be understood that the element of the ith row and the jth column in the similarity matrix and the element of the jth row and the jth column are the similarity between the ith object to be grouped and the jth object to be grouped. Therefore, in theory, any side of the diagonal in the similarity matrix includes the similarity between every two objects to be grouped, and the optimized object pair can be determined by using only the similarity at any side of the diagonal in the similarity matrix.
The diagonal of the similarity matrix refers herein to: a diagonal line is defined by the elements of row 1 and column 1 and the elements of column N. For example, taking the foregoing similarity matrix as an example, the terms < S1, S2>, < S1, S3>, < S1, S4>, < S2, S3>, < S2, S4>, < S3, S4> are all the similarities located on one side of the diagonal, while the terms < S2, S1>, < S3, S2>, < S4, S1>, < S4, S2>, < S4, S3> are all the similarities located on the other side of the diagonal.
The method for determining the optimizing object pair by using the similarity of any side of the diagonal line in the similarity matrix is as follows: and ordering all the similarities positioned on one side of the diagonal line in the similarity matrix according to the order from high to low to obtain a similarity sequence. And according to the sequence from front to back, aiming at each similarity in the similarity sequence, if each object to be grouped corresponding to the similarity does not belong to any optimizing object pair, taking two objects to be grouped corresponding to the similarity as a pair of optimizing object pairs until a preset termination condition is reached. Wherein the preset termination conditions include, but are not limited to, any of the following conditions: the number of the determined optimizing object pairs reaches a preset number threshold, the number of the loops reaches a preset number threshold, and the like.
Taking the foregoing similarity matrix as an example, assume that the termination condition is that the number of determined pairs of optimizing objects reaches 2, and that all the similarities on the diagonal side refer to: < S1, S2>, < S1, S3>, < S1, S4>, < S2, S3>, < S2, S4>, < S3, S4>, and assuming that < S1, S2> < S1, S3> < S1, S4> < S2, S3> < S2, S4> < S3, S4>, the resulting similarity sequence is { < S1, S2>, < S1, S3>, < S1, S4>, < S2, S3>, S4>, < S3, S4> }, first for < S1, S2>, since no pair of optimizing objects is initially determined, S1, S2 do not belong to any pair of optimizing objects, and then for < S1, S3>, since a pair of optimizing objects < S1, S2> has been determined, and then for < S1, S2> as a pair of optimizing objects, S1, S2, S4> and < S2, S4> as a pair of optimizing objects. And for < S3, S4>, since S3, S4 do not belong to any optimizing object pair, the < S3, S4> is taken as a pair of optimizing object pairs.
By adopting the embodiment, the intersection between different optimizing object pairs can be avoided, so that the intersection between the control group and the experimental group divided based on the optimizing object pairs does not exist, the mutual interference of the experiments of the control group and the experimental group is effectively avoided, and the reliability of the A/B experiment is improved.
The method of optimizing and sampling provided in the present disclosure will be described with reference to a specific example, referring to fig. 4, in which N objects to be grouped are present in total, and each object to be grouped is characterized by an M-dimensional feature vector.
Firstly, object features of each object to be grouped are spliced into an N-by-M-dimensional feature matrix, and elements of an ith row and a jth column in the feature matrix are as follows: the j-th feature value (hereinafter referred to as xij) of the object feature of the i-th object to be grouped. And inputting the characteristic matrix into a preset feedforward neural network.
In this example, the aforementioned full connection layer is located in a preset feedforward neural network, and the output of the preset feedforward neural network is the similarity between every two objects to be grouped. The input layers of the preset feedforward neural network are two layers. The first input layer includes n×m neurons, and the input of each neuron is one eigenvalue in the eigenvmatrix, and the inputs of different neurons are different. For convenience of description, neurons input as the feature value xij in the first input layer are denoted as neurons ij.
Each neuron in the first input layer corresponds to two neurons in the second input layer, which are respectively denoted as positive and negative neurons for convenience of description. For convenience of description, a positive neuron corresponding to the neuron ij is referred to as a positive neuron ij, and a negative neuron corresponding to the neuron ij is referred to as a negative neuron ij. The neuron ij is used to input xij to the positive neuron ij and-xij to the negative neuron ij.
In this example, the fully-connected layer includes N (N-1) M difference solution units, and for convenience of description, the difference solution unit corresponding to the eigenvalue pair { xij, xkj } is denoted as a difference solution unit ikj, where k is a positive integer with a value range of [1, N ], and k is not equal to i.
Then for the difference solving unit ikj, the inputs of the first difference solving unit are xij output by the positive neuron ij and-xkj output by the negative neuron kj. The inputs of the second difference solving unit are xkj output by the positive neuron kj and-xij output by the negative neuron ij.
And the first difference neuron is used for summing the inputs, namely calculating xij-xkj, inputting xij-xkj to the summing neuron if xij-xkj is larger than 0, and inputting 0 to the summing neuron if xij-xkj is not larger than 0.
And the second difference neuron is used for summing the inputs, namely calculating xkj-xij, inputting xkj-xij to the summing neuron if xkj-xij is larger than 0, and inputting 0 to the summing neuron if xkj-xij is not larger than 0.
And the summing neuron is used for summing the inputs and outputting the obtained result to the output layer. It will be appreciated that if xij > xkj, then the inputs to the summing neurons are xij-xkj and 0, and thus the result of the summation is xij-xkj, i.e., the absolute value of the difference between xij and xkj. Similarly, if xij < xkj > 0, the inputs to the summing neuron are 0 and xkj-xij, so that the result of the summation is xkj-xij, i.e., the absolute value of the difference between xij and xkj. It can be seen that in this example, the difference solving unit can accurately calculate the absolute value of the difference between the two eigenvalues in the eigenvalue pair.
In this example, the input of the output layer is the absolute value of the difference value output by each difference value solving unit, and the output layer is used for determining the similarity matrix according to the absolute value of the difference value. In this example, the similarity matrix is an n×n-dimensional matrix, and for the similarity matrix, reference may be made to the foregoing related description, which is not repeated herein.
All the similarities on the diagonal side of the similarity matrix are extracted and arranged side by side to obtain a similarity sequence as shown in fig. 4, in this example, the degree of similarity represented by the similarity is inversely related to the numerical value of the similarity, so that the numerical values of the similarities are sorted in order from small to large at the time of sorting.
And determining optimizing object pairs based on the similarity sequences, and dividing one object to be grouped in each optimizing pair into an experimental group and the other object to be grouped into a comparison group. For determining the optimizing object pair based on the similarity sequence, reference may be made to the foregoing related description, and will not be repeated herein.
Corresponding to the foregoing optimizing and sampling method, the disclosure further provides an optimizing and sampling device, as shown in fig. 5, including:
a feature acquisition module 501, configured to acquire object features of each of a plurality of objects to be grouped;
the difference solving module 502 is configured to weight and sum all the object features according to a feature value pair formed by feature values of the same feature dimension in every two object features, so as to obtain a difference absolute value of two feature values in the feature value pair;
A similarity solving module 503, configured to calculate a similarity between each two objects to be grouped according to the absolute value of the difference value;
the object pair screening module 504 is configured to select at least one pair of the objects to be grouped, where the similarity is greater than a preset similarity threshold, as an optimized object pair.
In a possible embodiment, the difference solving module 502 performs weighted summation on all the object features for a feature value pair composed of feature values of the same feature dimension in each two object features, to obtain an absolute value of a difference between two feature values in the feature value pair, including:
and inputting the characteristic value pairs into a preset full-connection layer aiming at characteristic value pairs formed by characteristic values of the same characteristic dimension in every two object characteristics to obtain the absolute difference value of the two characteristic values in the characteristic value pairs output by the full-connection layer.
The full connection layer is used for calculating the absolute value of the difference value of the two characteristic values in the characteristic value pair according to each characteristic value pair, and outputting the absolute value of the difference value.
In one possible embodiment, the fully connected layer includes a plurality of difference solving units;
the difference solving module 502 inputs the feature value pair to a preset full connection layer, and obtains the absolute difference value of two feature values in the feature value pair output by the full connection layer, including:
Inputting the characteristic value pairs to the difference value solving units corresponding to the characteristic value pairs to obtain the difference absolute values of two characteristic values in the characteristic value pairs output by the difference value solving units;
wherein different pairs of the eigenvalues correspond to different difference solving units, each of the difference solving units is configured to calculate a difference absolute value of two eigenvalues in the pairs of eigenvalues input to the difference solving unit, and output the difference absolute value.
In a possible embodiment, the difference solving unit includes a first difference neuron, a second difference neuron, and a summing neuron;
the difference value solving module 502 inputs the eigenvalue pair to the difference value solving unit corresponding to the eigenvalue pair, and obtains the absolute difference value of the two eigenvalues in the eigenvalue pair output by the difference value solving unit, including:
respectively inputting the characteristic value pairs into a first difference value neuron and a second difference value neuron in a difference value solving unit corresponding to the characteristic value pairs to obtain a first output value output by the first difference value neuron and a second output value output by the second difference value neuron;
Inputting the first output value and the second output value into a summation neuron in a difference value solving unit corresponding to the characteristic value pair to obtain an output value output by the summation neuron, wherein the output value is used as an absolute value of a difference value of two characteristic values in the characteristic value pair;
the first difference value neuron is used for subtracting one characteristic value from the other characteristic value in the characteristic value pair input to the difference value solving unit to obtain a first difference value; outputting the first difference value if the first difference value is greater than 0, and outputting 0 if the first difference value is not greater than 0;
the second difference value neuron is configured to subtract the one feature value from the other feature value to obtain a second difference value; outputting the second difference value if the second difference value is larger than 0, and outputting 0 if the second difference value is not larger than 0;
and the summation neuron is used for calculating the summation result of the first output value and the second output value and outputting the summation result.
In one possible embodiment, the similarity solving module 503 calculates, according to the absolute value of the difference, a similarity between each two objects to be grouped, including:
And generating a similarity matrix according to the absolute value of the difference, wherein the elements of the ith row and the jth column of the similarity matrix are the similarity between the ith object to be grouped and the jth object to be grouped, i and j are positive integers with the value range of [1, N ], and N is the number of the objects to be grouped.
In a possible embodiment, the object pair filtering module 504 selects, according to the similarity, at least one pair of objects to be grouped that satisfy a preset filtering condition as a optimizing object pair, including:
sequencing all the similarities positioned on any side of a diagonal line in the similarity matrix according to the sequence from high to low to obtain a similarity sequence;
and according to the sequence from front to back, aiming at each similarity in the similarity sequence in sequence, if each object to be grouped corresponding to the similarity does not belong to any optimizing object pair, taking two objects to be grouped corresponding to the similarity as a pair of optimizing object pairs until a preset termination condition is reached.
In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.
It should be noted that, in some application scenarios, the object to be grouped in the embodiments of the present disclosure may be a head model, and the head model in the embodiments is not a head model for a specific user, and cannot reflect personal information of a specific user.
It should be noted that, in other application scenarios, the objects to be grouped in the embodiments of the present disclosure may be two-dimensional face images, and the two-dimensional face images in the embodiments are from a public dataset.
According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
Fig. 6 illustrates a schematic block diagram of an example electronic device 600 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 6, the apparatus 600 includes a computing unit 601 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 602 or a computer program loaded from a storage unit 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data required for the operation of the device 600 may also be stored. The computing unit 601, ROM 602, and RAM603 are connected to each other by a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
Various components in the device 600 are connected to the I/O interface 605, including: an input unit 606 such as a keyboard, mouse, etc.; an output unit 607 such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, optical disk, or the like; and a communication unit 609 such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The computing unit 601 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 601 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as method XXX. For example, in some embodiments, method XXX may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 600 via the ROM 602 and/or the communication unit 609. One or more of the steps of method XXX described above may be performed when a computer program is loaded into RAM603 and executed by computing unit 601. Alternatively, in other embodiments, computing unit 601 may be configured to perform method XXX by any other suitable means (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein.
The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (14)
1. An optimizing sampling method, comprising:
acquiring respective object characteristics of a plurality of objects to be grouped; the object to be grouped is a tester; the object features are data features generated by the objects to be grouped within a preset time length;
for the characteristic value pairs formed by characteristic values of the same characteristic dimension in every two object characteristics, carrying out weighted summation on all the object characteristics to obtain the difference absolute values of the two characteristic values in the characteristic value pairs;
Calculating the similarity between every two objects to be grouped according to the absolute value of the difference value;
selecting at least one pair of objects to be grouped, wherein the similarity is larger than a preset similarity threshold value, as an optimizing object pair; and determining one tester in the optimizing object pair as a control group tester, and determining the other tester in the optimizing object pair as an experimental group tester to acquire an application without a new function used by the control group tester, wherein the experimental group tester uses an experimental result with the new function.
2. The method according to claim 1, wherein the step of weighting and summing all the object features for the feature value pairs composed of feature values of the same feature dimension in each two of the object features to obtain a difference absolute value between two feature values in the feature value pairs includes:
inputting the characteristic value pairs into a preset full-connection layer aiming at characteristic value pairs formed by characteristic values of the same characteristic dimension in every two object characteristics to obtain the absolute value of the difference value of the two characteristic values in the characteristic value pairs output by the full-connection layer;
the full connection layer is used for calculating the absolute value of the difference value of the two characteristic values in the characteristic value pair according to each characteristic value pair, and outputting the absolute value of the difference value.
3. The method of claim 2, wherein the fully connected layer comprises a plurality of difference solving units;
inputting the characteristic value pair to a preset full-connection layer to obtain the absolute value of the difference value of two characteristic values in the characteristic value pair output by the full-connection layer, wherein the method comprises the following steps:
inputting the characteristic value pairs to the difference value solving units corresponding to the characteristic value pairs to obtain the difference absolute values of two characteristic values in the characteristic value pairs output by the difference value solving units;
wherein different pairs of the eigenvalues correspond to different difference solving units, each of the difference solving units is configured to calculate a difference absolute value of two eigenvalues in the pairs of eigenvalues input to the difference solving unit, and output the difference absolute value.
4. The method of claim 3, wherein the difference solving unit comprises a first difference neuron, a second difference neuron, and a summing neuron;
the step of inputting the characteristic value pair to the difference value solving unit corresponding to the characteristic value pair to obtain the absolute value of the difference value of the two characteristic values in the characteristic value pair output by the difference value solving unit, comprising the following steps:
Respectively inputting the characteristic value pairs into a first difference value neuron and a second difference value neuron in a difference value solving unit corresponding to the characteristic value pairs to obtain a first output value output by the first difference value neuron and a second output value output by the second difference value neuron;
inputting the first output value and the second output value into a summation neuron in a difference value solving unit corresponding to the characteristic value pair to obtain an output value output by the summation neuron, wherein the output value is used as an absolute value of a difference value of two characteristic values in the characteristic value pair;
the first difference value neuron is used for subtracting one characteristic value from the other characteristic value in the characteristic value pair input to the difference value solving unit to obtain a first difference value; outputting the first difference value if the first difference value is greater than 0, and outputting 0 if the first difference value is not greater than 0;
the second difference value neuron is configured to subtract the one feature value from the other feature value to obtain a second difference value; outputting the second difference value if the second difference value is larger than 0, and outputting 0 if the second difference value is not larger than 0;
and the summation neuron is used for calculating the summation result of the first output value and the second output value and outputting the summation result.
5. The method of claim 1, wherein the calculating the similarity between each two objects to be grouped according to the absolute value of the difference value comprises:
and generating a similarity matrix according to the absolute value of the difference, wherein the elements of the ith row and the jth column of the similarity matrix are the similarity between the ith object to be grouped and the jth object to be grouped, i and j are positive integers with the value range of [1, N ], and N is the number of the objects to be grouped.
6. The method according to claim 5, wherein selecting at least one pair of the objects to be grouped satisfying a preset screening condition as an optimized object pair according to the similarity comprises:
ordering all the similarities positioned on any side of a diagonal line in the similarity matrix according to the order from high to low to obtain a similarity sequence;
and according to the sequence from front to back, aiming at each similarity in the similarity sequence in sequence, if each object to be grouped corresponding to the similarity does not belong to any optimizing object pair, taking two objects to be grouped corresponding to the similarity as a pair of optimizing object pairs until a preset termination condition is reached.
7. An optimizing and sampling device, comprising:
The characteristic acquisition module is used for acquiring the object characteristics of each of a plurality of objects to be grouped; the object to be grouped is a tester; the object features are data features generated by the objects to be grouped within a preset time length;
the difference solving module is used for carrying out weighted summation on all the object features aiming at the feature value pairs formed by the feature values of the same feature dimension in every two object features to obtain the difference absolute values of the two feature values in the feature value pairs;
the similarity solving module is used for calculating the similarity between every two objects to be grouped according to the absolute value of the difference value;
the object pair screening module is used for selecting at least one pair of objects to be grouped, the similarity of which is greater than a preset similarity threshold value, as an optimizing object pair; and determining one tester in the optimizing object pair as a control group tester, and determining the other tester in the optimizing object pair as an experimental group tester to acquire an application without a new function used by the control group tester, wherein the experimental group tester uses an experimental result with the new function.
8. The apparatus of claim 7, wherein the difference solution module performs weighted summation on all the object features for a feature value pair composed of feature values of the same feature dimension in each two object features, to obtain an absolute difference value of two feature values in the feature value pair, and includes:
Inputting the characteristic value pairs into a preset full-connection layer aiming at characteristic value pairs formed by characteristic values of the same characteristic dimension in every two object characteristics to obtain the absolute value of the difference value of the two characteristic values in the characteristic value pairs output by the full-connection layer;
the full connection layer is used for calculating the absolute value of the difference value of the two characteristic values in the characteristic value pair according to each characteristic value pair, and outputting the absolute value of the difference value.
9. The apparatus of claim 8, wherein the fully connected layer comprises a plurality of difference solving units;
the difference value solving module inputs the characteristic value pair to a preset full-connection layer to obtain the absolute value of the difference value of two characteristic values in the characteristic value pair output by the full-connection layer, and the difference value solving module comprises the following steps:
inputting the characteristic value pairs to the difference value solving units corresponding to the characteristic value pairs to obtain the difference absolute values of two characteristic values in the characteristic value pairs output by the difference value solving units;
wherein different pairs of the eigenvalues correspond to different difference solving units, each of the difference solving units is configured to calculate a difference absolute value of two eigenvalues in the pairs of eigenvalues input to the difference solving unit, and output the difference absolute value.
10. The apparatus of claim 9, wherein the difference solving unit comprises a first difference neuron, a second difference neuron, and a summing neuron;
the difference value solving module inputs the characteristic value pair to the difference value solving unit corresponding to the characteristic value pair to obtain the difference absolute value of two characteristic values in the characteristic value pair output by the difference value solving unit, and the difference absolute value comprises:
respectively inputting the characteristic value pairs into a first difference value neuron and a second difference value neuron in a difference value solving unit corresponding to the characteristic value pairs to obtain a first output value output by the first difference value neuron and a second output value output by the second difference value neuron;
inputting the first output value and the second output value into a summation neuron in a difference value solving unit corresponding to the characteristic value pair to obtain an output value output by the summation neuron, wherein the output value is used as an absolute value of a difference value of two characteristic values in the characteristic value pair;
the first difference value neuron is used for subtracting one characteristic value from the other characteristic value in the characteristic value pair input to the difference value solving unit to obtain a first difference value; outputting the first difference value if the first difference value is greater than 0, and outputting 0 if the first difference value is not greater than 0;
The second difference value neuron is configured to subtract the one feature value from the other feature value to obtain a second difference value; outputting the second difference value if the second difference value is larger than 0, and outputting 0 if the second difference value is not larger than 0;
and the summation neuron is used for calculating the summation result of the first output value and the second output value and outputting the summation result.
11. The apparatus of claim 7, wherein the similarity solving module calculates a similarity between each two of the objects to be grouped according to the absolute difference value, comprising:
and generating a similarity matrix according to the absolute value of the difference, wherein the elements of the ith row and the jth column of the similarity matrix are the similarity between the ith object to be grouped and the jth object to be grouped, i and j are positive integers with the value range of [1, N ], and N is the number of the objects to be grouped.
12. The apparatus of claim 11, wherein the object pair filtering module selects, as the optimizing object pair, at least one pair of the objects to be grouped that satisfy a preset filtering condition according to the similarity, the optimizing object pair comprising:
ordering all the similarities positioned on any side of a diagonal line in the similarity matrix according to the order from high to low to obtain a similarity sequence;
And according to the sequence from front to back, aiming at each similarity in the similarity sequence in sequence, if each object to be grouped corresponding to the similarity does not belong to any optimizing object pair, taking two objects to be grouped corresponding to the similarity as a pair of optimizing object pairs until a preset termination condition is reached.
13. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
14. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210349040.1A CN114781496B (en) | 2022-04-01 | 2022-04-01 | Optimizing sampling method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210349040.1A CN114781496B (en) | 2022-04-01 | 2022-04-01 | Optimizing sampling method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114781496A CN114781496A (en) | 2022-07-22 |
| CN114781496B true CN114781496B (en) | 2023-11-07 |
Family
ID=82428200
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210349040.1A Active CN114781496B (en) | 2022-04-01 | 2022-04-01 | Optimizing sampling method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114781496B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522921A (en) * | 2018-09-18 | 2019-03-26 | 义语智能科技（上海）有限公司 | Statement similarity method of discrimination and equipment |
| CN111309994A (en) * | 2020-01-22 | 2020-06-19 | 北京三快在线科技有限公司 | User matching method and device, electronic equipment and readable storage medium |
| CN111353052A (en) * | 2020-02-17 | 2020-06-30 | 北京达佳互联信息技术有限公司 | Multimedia object recommendation method and device, electronic equipment and storage medium |
| CN111428133A (en) * | 2020-03-19 | 2020-07-17 | 腾讯科技（北京）有限公司 | Artificial intelligence based recommendation method and device, electronic equipment and storage medium |
| CN112687257A (en) * | 2021-03-11 | 2021-04-20 | 北京新唐思创教育科技有限公司 | Sentence similarity judging method and device, electronic equipment and readable storage medium |
| CN112907128A (en) * | 2021-03-23 | 2021-06-04 | 百度在线网络技术（北京）有限公司 | Data analysis method, device, equipment and medium based on AB test result |
| CN113313192A (en) * | 2021-06-15 | 2021-08-27 | 成都恒创新星科技有限公司 | Image similarity judgment method and system |
| WO2021223165A1 (en) * | 2020-05-07 | 2021-11-11 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for object evaluation |
| CN113742600A (en) * | 2021-11-05 | 2021-12-03 | 北京达佳互联信息技术有限公司 | Resource recommendation method and device, computer equipment and medium |
-
2022
- 2022-04-01 CN CN202210349040.1A patent/CN114781496B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522921A (en) * | 2018-09-18 | 2019-03-26 | 义语智能科技（上海）有限公司 | Statement similarity method of discrimination and equipment |
| CN111309994A (en) * | 2020-01-22 | 2020-06-19 | 北京三快在线科技有限公司 | User matching method and device, electronic equipment and readable storage medium |
| CN111353052A (en) * | 2020-02-17 | 2020-06-30 | 北京达佳互联信息技术有限公司 | Multimedia object recommendation method and device, electronic equipment and storage medium |
| CN111428133A (en) * | 2020-03-19 | 2020-07-17 | 腾讯科技（北京）有限公司 | Artificial intelligence based recommendation method and device, electronic equipment and storage medium |
| WO2021223165A1 (en) * | 2020-05-07 | 2021-11-11 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for object evaluation |
| CN112687257A (en) * | 2021-03-11 | 2021-04-20 | 北京新唐思创教育科技有限公司 | Sentence similarity judging method and device, electronic equipment and readable storage medium |
| CN112907128A (en) * | 2021-03-23 | 2021-06-04 | 百度在线网络技术（北京）有限公司 | Data analysis method, device, equipment and medium based on AB test result |
| CN113313192A (en) * | 2021-06-15 | 2021-08-27 | 成都恒创新星科技有限公司 | Image similarity judgment method and system |
| CN113742600A (en) * | 2021-11-05 | 2021-12-03 | 北京达佳互联信息技术有限公司 | Resource recommendation method and device, computer equipment and medium |
Non-Patent Citations (2)
| Title |
|---|
| Deep Spatio-Temporal Neural Networks for Click-Through Rate Prediction;Wentao Ouyang 等;《arXiv》;1-9 * |
| 基于综合相似度和社交标签的推荐算法;时念云 等;《计算机系统应用》;第26卷(第10期);178-183 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114781496A (en) | 2022-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11295208B2 (en) | Robust gradient weight compression schemes for deep learning applications | |
| Lou et al. | Predicting network controllability robustness: A convolutional neural network approach | |
| CN113313053B (en) | Image processing method, device, apparatus, medium, and program product | |
| EP4343616A1 (en) | Image classification method, model training method, device, storage medium, and computer program | |
| CN117041017B (en) | Intelligent operation and maintenance management method and system for data center | |
| CN114490065A (en) | Load prediction method, device and equipment | |
| CN110337636A (en) | Data transfer device and device | |
| CN115374948A (en) | Quantum neural network training method, data processing method, device and medium | |
| CN113627361B (en) | Training method and device for face recognition model and computer program product | |
| CN114693934A (en) | Training method of semantic segmentation model, video semantic segmentation method and device | |
| CN113486302A (en) | Data processing method and device | |
| CN114781496B (en) | Optimizing sampling method and device and electronic equipment | |
| CN113642654B (en) | Image feature fusion method and device, electronic equipment and storage medium | |
| CN113516185B (en) | Model training method, device, electronic equipment and storage medium | |
| CN112749707A (en) | Method, apparatus, and medium for object segmentation using neural networks | |
| CN111709784A (en) | Method, apparatus, device and medium for generating user retention time | |
| JP2020030702A (en) | Learning device, learning method, and learning program | |
| CN113361402B (en) | Training method of recognition model, method, device and equipment for determining accuracy | |
| CN115456167B (en) | Lightweight model training method, image processing device and electronic equipment | |
| Hayasaka et al. | On the least square error and prediction square error of function representation with discrete variable basis | |
| CN117522883A (en) | Quantum computation-based image processing method, device, electronic equipment and medium | |
| CN115171225A (en) | Image detection method and training method of image detection model | |
| CN116436828A (en) | Network traffic prediction and model training method, device, equipment and storage medium | |
| CN114943995A (en) | Training method of face recognition model, face recognition method and device | |
| CN117617951A (en) | Human behavior recognition method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |