CN110334806A - A kind of confrontation sample generating method based on production confrontation network - Google Patents
A kind of confrontation sample generating method based on production confrontation network Download PDFInfo
- Publication number
- CN110334806A CN110334806A CN201910459852.XA CN201910459852A CN110334806A CN 110334806 A CN110334806 A CN 110334806A CN 201910459852 A CN201910459852 A CN 201910459852A CN 110334806 A CN110334806 A CN 110334806A
- Authority
- CN
- China
- Prior art keywords
- sample
- loss function
- generator
- confrontation
- network
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 11
- 230000002708 enhancing effect Effects 0.000 claims abstract description 10
- 230000003044 adaptive effect Effects 0.000 claims abstract description 5
- 238000012360 testing method Methods 0.000 claims description 21
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 7
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 6
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000005012 migration Effects 0.000 abstract description 5
- 238000013508 migration Methods 0.000 abstract description 5
- 230000008485 antagonism Effects 0.000 abstract 1
- 230000009977 dual effect Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 34
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 235000000332 black box Nutrition 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of confrontation sample generating methods based on production confrontation network, including generator G, discriminator D, spatial alternation module ST and target classification network F, generator G generates disturbance, disturbance is added in original sample the sample that creates antagonism, generator G is trained further according to the loss function of discriminator D and target classification network F, trained generator G is finally obtained, it is adaptive to resisting sample to be that different input samples generate using trained generator G.The present invention fights network using production, is embedded in the enhancing module based on spatial alternation, carries out dual training using unsupervised mode, improves the generalization ability and robustness of challenge model, and then enhance the migration and robustness to resisting sample.
Description
Technical field
The present invention relates to machine learning field, more particularly, to it is a kind of based on production confrontation network to resisting sample
Generation method.
Background technique
It is a hot issue of current machine learning areas research to attack resistance.Principle to attack resistance is to pass through confrontation
Sample (new samples that addition is obtained by the not noticeable small sample perturbations of meticulously trained human eye into former data sample) is taken advantage of
Deep neural network is deceived, its judgement that makes mistake is made.
Most of attack algorithms (such as based on gradient and based on the method for optimization) based on deep neural network are all at present
For test process or test data set, and need to carry out whitepack access to the architecture of model and parameter always (such as to obtain
Gradient relevant to input is taken just to need to know the weight of target network).But current deep learning system is usually for peace
Full reason does not allow to carry out whitepack access to model, only allows to carry out queried access to model, i.e., model is regarded as black box.For
The attack of such case is referred to as black box attack, but the success rate of current most of black box attacks is not high, because most of black
Box attack method is all based on the transportable property (Transferability) to resisting sample.Transportable property is one to resisting sample
A common properties, referring to also has good attack effect to other scopes of a variable to resisting sample according to what finite sample generated.
In the black box attack that can not obtain Destination Network Structure and training dataset, transportable property is most important.How
It efficiently generates that transportable property is strong, attack performance is stable to resisting sample, is an extremely significant and extremely challenging problem.
In conclusion although existing research demonstrates existing attack method in the identical structure nerve net of different data training
There is certain migration between network and between the different structure neural network of same task training, such as document [1]
Goodfellow I J,Shlens J,Szegedy C,et al.Explaining and Harnessing Adversarial
Examples [J] .International Conference on Learning Representations, 2015, document [2]
Kurakin A,Goodfellow I J,Bengio S,et al.Adversarial examples in the physical
World [J] .arXiv:Computer Vision and Pattern Recognition, 2017, document [3]
Moosavidezfooli S,Fawzi A,Frossard P,et al.DeepFool:A Simple and Accurate
Method to Fool Deep Neural Networks[J].Computer Vision and Pattern
Recognition, 2016:2574-2582 and document [4] Xiao C, Li B, Zhu J Y, et al.Generating
Adversarial Examples with Adversarial Networks[J].2018；But there are still confrontation sample excessively according to
The problems such as relying object module and causing property transportable to resisting sample poor, success attack rate is low, attack efficiency is low.
Summary of the invention
The present invention provides a kind of confrontation sample generating method based on production confrontation network, is embodied as different input samples
Adaptive generate has transportable property and robust sex resistance sample.
In order to solve the above technical problems, technical scheme is as follows:
A kind of confrontation sample generating method based on production confrontation network, comprising the following steps:
S1: original sample x is inputted in generator G, the loss function of the generator G output disturbance G (x), generator G is
LG, disturbance G (x), which is added in original sample x, to be obtained to resisting sample x '=x+G (x), the target of generator different from general GAN
It is to generate disturbance rather than final image, i.e., output image is equal to the output image that input picture adds generator G, pair of generation
The details and texture of resisting sample are replicated from input picture, and the details of original image is greatly remained；
S2: by S1 obtain in resisting sample x ' input discriminator D, the discriminator D is distinguished to resisting sample x ' and as former state
This x obtains the loss function L of discriminator DD；
S3: being input in enhancing module ST resisting sample x ' for what S1 was obtained, and the enhancing module ST is based on spatial alternation,
To resisting sample x ' carry out spatial alternation operation, enhancing module ST output, treated to resisting sample x ' by affine transformationst=
Tθ(x+G (x)), T in formulaθFor transforming function transformation function；
S4: passing through affine transformation, treated to resisting sample x 'stObject-class model F is inputted, target classification mould is obtained
The loss function L of type FF；
S5: being L according to the loss function of generator GG, discriminator D loss function LDWith the loss of object-class model F
Function LFConstruct objective function LGANFor training challenge model GAN, trained generator G is obtained；
S6: it is adaptive to resisting sample to be that different input samples generate using trained generator G.
Preferably, the loss function of generator G uses L2 norm to lose as distance metric, is specifically expressed as follows:
LG=max (0, | | G (x) | |2-c)
Wherein, c is customized constant, it allows a user to specify the disturbance quantity of addition, and it is various right to generate
Resisting sample can facilitate a better understanding of the feature space to resisting sample.The loss can also stablize the training of GAN.
Preferably, the discriminator D is binary neural network classifier.
Preferably, the loss function of the discriminator D specifically:
LD=logD (x)+log (1-D (x+G (x))).
Preferably, the loss function of object-class model F is in the attack for having target are as follows:
LF=L (F (Tθ(x+G(x))),y′)
Indicate prediction the distance between class and target class y ', in formula, L is cross entropy loss function；
The loss function of object-class model F is in aimless attack are as follows:
LF=-L (F (Tθ(x+G(x))),y)
Indicate the negative distance between prediction class and original tag class y, in formula, L is cross entropy loss function.
Preferably, objective function LGANIt indicates are as follows:
LGAN=LF+αLD+β·LG
In formula, α and β are constants, for controlling the relative importance of each objective function, LGFor generating small sample perturbations, LDWith
Resisting sample is shown similar to original sample in encourage to generate, and LFFor optimizing to resisting sample, success attack rate is improved, is passed through
It minimizes generator loss function and maximizes discriminator loss function argmingmaxdLGANSolution obtains generator G and discriminator
D。
Preferably, step S5 further include trained generator G is tested, specifically includes the following steps:
S5.1: generating disturbance using trained generator G, to generate test to resisting sample, will test to resisting sample
The target classification network for inputting different structure, makes its classification error；
S5.2: carrying out spatial alternation to resisting sample to the test of S5.1, generates new test to resisting sample, new test confrontation sample
This input target classification network, makes its classification error.
Compared with prior art, the beneficial effect of technical solution of the present invention is:
Compared with existing attack algorithm, method proposed by the present invention, which does not need access original object disaggregated model, to be had
Effect ground is that different input samples generate attack sample, and inquiry is high with formation efficiency, challenge model generalization ability and strong robustness, energy
The transportable property and robustness to resisting sample are effectively improved, and then improves black box success attack rate.The applicability of the method for the present invention
Relatively wide, versatility is stronger, and the success attack rate on the model of different types of data set and different structure is all higher.
Detailed description of the invention
Fig. 1 is a kind of confrontation sample generating method flow chart that network is fought based on production.
Fig. 2 is a kind of confrontation sample generating method model schematic that network is fought based on production, and dotted line represents in figure
Training process, solid line represent test process.
Fig. 3 is black box attack robust implementation flow chart.
Specific embodiment
The attached figures are only used for illustrative purposes and cannot be understood as limitating the patent；
In order to better illustrate this embodiment, the certain components of attached drawing have omission, zoom in or out, and do not represent actual product
Size；
To those skilled in the art, it is to be understood that certain known features and its explanation, which may be omitted, in attached drawing
's.
The following further describes the technical solution of the present invention with reference to the accompanying drawings and examples.
Embodiment 1
A kind of confrontation sample generating method based on production confrontation network, such as Fig. 1 to 2, comprising the following steps:
S1: original sample x is inputted in generator G, the loss function of the generator G output disturbance G (x), generator G is
LG, disturbance G (x), which is added in original sample x, to be obtained to resisting sample x '=x+G (x), the target of generator different from general GAN
It is to generate disturbance rather than final image, i.e., output image is equal to the output image that input picture adds generator G, pair of generation
The details and texture of resisting sample are replicated from input picture, and the details of original image, the loss of generator G are greatly remained
Function uses L2 norm to lose as distance metric, is specifically expressed as follows:
LG=max (0, | | G (x) | |2-c)
Wherein, c is customized constant；
S2: by S1 obtain in resisting sample x ' input discriminator D, the discriminator D is distinguished to resisting sample x ' and as former state
This x, the discriminator D are binary neural network classifier, obtain the loss function L of discriminator DD=logD (x)+log (1-D
(x+G(x)))；
S3: being input in enhancing module ST resisting sample x ' for what S1 was obtained, and the enhancing module ST is based on spatial alternation,
To resisting sample x ' carry out spatial alternation operation, enhancing module ST output, treated to resisting sample x ' by affine transformationst=
Tθ(x+G (x)), T in formulaθFor transforming function transformation function；
S4: passing through affine transformation, treated to resisting sample x 'stObject-class model F is inputted, target classification mould is obtained
The loss function L of type FF, the loss function of object-class model F is in the attack for having target are as follows:
LF=L (F (Tθ(x+G(x))),y′)
Indicate prediction the distance between class and target class y ', in formula, L is cross entropy loss function；
The loss function of object-class model F is in aimless attack are as follows:
LF=-L (F (Tθ(x+G(x))),y)
Indicate the negative distance between prediction class and original tag class y, in formula, L is cross entropy loss function；
S5: being L according to the loss function of generator GG, discriminator D loss function LDWith the loss of object-class model F
Function LFConstruct objective function LGANFor training challenge model GAN, trained generator G and discriminator D, target letter are obtained
Number LGANIt indicates are as follows:
LGAN=LF+αLD+β·LG
In formula, α and β are constants, for controlling the relative importance of each objective function, LGFor generating small sample perturbations, LDWith
Resisting sample is shown similar to original sample in encourage to generate, and LFFor optimizing to resisting sample, success attack rate is improved；
Further include trained generator G is tested, specifically includes the following steps:
S5.1: generating disturbance using trained generator G, to generate test to resisting sample, will test to resisting sample
The target classification network for inputting different structure, makes its classification error；
S5.2: carrying out spatial alternation to resisting sample to the test of S5.1, generates new test to resisting sample, new test confrontation sample
This input target classification network, makes its classification error.
S6: it is adaptive to resisting sample to be that different input samples generate using trained generator G.
In the specific implementation process, robustness test is carried out so that black box is attacked as an example, detailed process is as shown in Figure 3.
1) target of attack F is selected.It is utilized using the training of CIFAR-10 data set ResNet-18, ResNet-34 and VGG-16
GTSRB data set trains VGG-16 and Multi-Scale CNN, obtain two groups totally five target of attack model F=F1, F2,
F3,F4,F5}.Wherein, ResNet-34 and VGG-16 is respectively as test to the ash box and black-box model of resisting sample.
2) data prediction.In order to exclude the influence of classification error caused by the performance of network itself, by target classification
The screening sample that network can correctly classify comes out, as the original sample generated to resisting sample.
3) it generates to resisting sample.Training process according to fig. 2 generates challenge model, and using its generation to resisting sample.
4) validity to resisting sample is tested.If what is generated can successfully cheat target classification network F to resisting sample and make it
Classification error illustrates that the present embodiment attack method is effective.
5) the transportable property to resisting sample is tested.If what is generated can cheat the target point of different structure to resisting sample simultaneously
Class network F1 and F2, make its classification error, then explanation is strong to resisting sample migration, conversely, then explanation is poor to resisting sample migration.
Compared with the generation of FGSM, BIM, DeepFool and advGAN method is to resisting sample, the success rate of attack is improved, then illustrates this reality
Migration to resisting sample can be effectively improved by applying a method.
Test the robustness to resisting sample.Spatial alternation is carried out to resisting sample to what step 3) generated, generates new confrontation sample
This, still can successfully cheat the target classification network F2 of step 1), then illustrate generate to resisting sample strong robustness.With FGSM,
BIM, DeepFool compare resisting sample with what advGAN method generated, improve the success rate of attack, then illustrate present implementation
The robustness to resisting sample can be effectively improved.
CIFAR-10 data set experimental result is as shown in table 1:
Table 1
GTSRB data set experimental result is as shown in table 2:
Table 2
The same or similar label correspond to the same or similar components；
The terms describing the positional relationship in the drawings are only for illustration, should not be understood as the limitation to this patent；
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention.For those of ordinary skill in the art, may be used also on the basis of the above description
To make other variations or changes in different ways.There is no necessity and possibility to exhaust all the enbodiments.It is all this
Made any modifications, equivalent replacements, and improvements etc., should be included in the claims in the present invention within the spirit and principle of invention
Protection scope within.
Claims (7)
1. a kind of confrontation sample generating method based on production confrontation network, which comprises the following steps:
S1: original sample x being inputted in generator G, the generator G output disturbance G (x), and the loss function of generator G is LG, disturb
Dynamic G (x), which is added in original sample x, to be obtained to resisting sample x '=x+G (x)；
S2: by S1 obtain in resisting sample x ' input discriminator D, the discriminator D is distinguished to resisting sample x ' and original sample x,
Obtain the loss function L of discriminator DD；
S3: resisting sample x ' is input in enhancing module ST by what S1 was obtained, the enhancing module ST is based on spatial alternation, to right
Resisting sample x ' carry out spatial alternation operation, by affine transformation, treated to resisting sample x ' for enhancing module ST outputst=Tθ(x+
G (x)), T in formulaθFor transforming function transformation function；
S4: passing through affine transformation, treated to resisting sample x 'stObject-class model F is inputted, obtains object-class model F's
Loss function LF；
S5: being L according to the loss function of generator GG, discriminator D loss function LDWith the loss function of object-class model F
LFConstruct objective function LGANFor training challenge model GAN, trained generator G is obtained；
S6: it is adaptive to resisting sample to be that different input samples generate using trained generator G.
2. the confrontation sample generating method according to claim 1 based on production confrontation network, which is characterized in that generate
The loss function of device G uses L2 norm to lose as distance metric, is specifically expressed as follows:
LG=max (0, | | G (x) | |2-c)
Wherein, c is customized constant.
3. the confrontation sample generating method according to claim 2 based on production network, which is characterized in that the identification
Device D is binary neural network classifier.
4. the confrontation sample generating method according to claim 3 based on production confrontation network, which is characterized in that described
The loss function of discriminator D specifically:
LD=log D (x)+log (1-D (x+G (x))).
5. the confrontation sample generating method according to claim 4 based on production confrontation network, which is characterized in that target
The loss function of disaggregated model F is in the attack for having target are as follows:
LF=L (F (Tθ(x+G(x))),y′)
Indicate prediction the distance between class and target class y ', in formula, L is cross entropy loss function；
The loss function of object-class model F is in aimless attack are as follows:
LF=-L (F (Tθ(x+G(x))),y)
Indicate the negative distance between prediction class and original tag class y, in formula, L is cross entropy loss function.
6. the confrontation sample generating method according to claim 5 based on production confrontation network, which is characterized in that target
Function LGANIt indicates are as follows:
LGAN=LF+αLD+β·LG
In formula, α and β are constants, maximize discriminator loss function argmin by minimizing generator loss functiongmaxdLGAN
Solution obtains generator G and discriminator D.
7. the confrontation sample generating method according to claim 6 based on production confrontation network, which is characterized in that step
S5 further include trained generator G is tested, specifically includes the following steps:
S5.1: generating disturbance using trained generator G, to generate test to resisting sample, test inputs resisting sample
The target classification network of different structure, makes its classification error；
S5.2: carrying out spatial alternation to resisting sample to the test of S5.1, generates new test to resisting sample, new test is defeated to resisting sample
Enter target classification network, makes its classification error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910459852.XA CN110334806A (en) | 2019-05-29 | 2019-05-29 | A kind of confrontation sample generating method based on production confrontation network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910459852.XA CN110334806A (en) | 2019-05-29 | 2019-05-29 | A kind of confrontation sample generating method based on production confrontation network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110334806A true CN110334806A (en) | 2019-10-15 |
Family
ID=68140522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910459852.XA Pending CN110334806A (en) | 2019-05-29 | 2019-05-29 | A kind of confrontation sample generating method based on production confrontation network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110334806A (en) |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110768971A (en) * | 2019-10-16 | 2020-02-07 | 伍军 | Confrontation sample rapid early warning method and system suitable for artificial intelligence system |
| CN111163472A (en) * | 2019-12-30 | 2020-05-15 | 浙江工业大学 | Signal identification attack defense method based on generative countermeasure network |
| CN111160217A (en) * | 2019-12-25 | 2020-05-15 | 中山大学 | Method and system for generating confrontation sample of pedestrian re-identification system |
| CN111210002A (en) * | 2019-12-30 | 2020-05-29 | 北京航空航天大学 | Multi-layer academic network community discovery method and system based on generation of confrontation network model |
| CN111241287A (en) * | 2020-01-16 | 2020-06-05 | 支付宝(杭州)信息技术有限公司 | Training method and device for generating generation model of confrontation text |
| CN111275115A (en) * | 2020-01-20 | 2020-06-12 | 星汉智能科技股份有限公司 | Method for generating counterattack sample based on generation counternetwork |
| CN111310802A (en) * | 2020-01-20 | 2020-06-19 | 星汉智能科技股份有限公司 | Anti-attack defense training method based on generation of anti-network |
| CN111340066A (en) * | 2020-02-10 | 2020-06-26 | 电子科技大学 | Confrontation sample generation method based on geometric vector |
| CN111539184A (en) * | 2020-04-29 | 2020-08-14 | 上海眼控科技股份有限公司 | Text data manufacturing method and device based on deep learning, terminal and storage medium |
| CN111738374A (en) * | 2020-08-28 | 2020-10-02 | 北京智源人工智能研究院 | Multi-sample anti-disturbance generation method and device, storage medium and computing equipment |
| CN111818101A (en) * | 2020-09-09 | 2020-10-23 | 平安国际智慧城市科技股份有限公司 | Network security detection method and device, computer equipment and storage medium |
| CN111898645A (en) * | 2020-07-03 | 2020-11-06 | 贵州大学 | Movable sample attack resisting method based on attention mechanism |
| CN111967592A (en) * | 2020-07-09 | 2020-11-20 | 中国电子科技集团公司第三十六研究所 | Method for generating counterimage machine recognition based on positive and negative disturbance separation |
| CN111967584A (en) * | 2020-08-19 | 2020-11-20 | 北京字节跳动网络技术有限公司 | Method, device, electronic equipment and computer storage medium for generating countermeasure sample |
| CN112162515A (en) * | 2020-10-10 | 2021-01-01 | 浙江大学 | Anti-attack method for process monitoring system |
| CN112818407A (en) * | 2021-04-16 | 2021-05-18 | 中国工程物理研究院计算机应用研究所 | Video privacy protection method based on generation countermeasure network |
| CN112884143A (en) * | 2019-11-29 | 2021-06-01 | 北京四维图新科技股份有限公司 | Method for training robust deep neural network model |
| CN113158190A (en) * | 2021-04-30 | 2021-07-23 | 河北师范大学 | Malicious code countermeasure sample automatic generation method based on generation type countermeasure network |
| CN113177599A (en) * | 2021-05-10 | 2021-07-27 | 南京信息工程大学 | Enhanced sample generation method based on GAN |
| CN113222480A (en) * | 2021-06-11 | 2021-08-06 | 支付宝(杭州)信息技术有限公司 | Training method and device for confrontation sample generation model |
| CN113361594A (en) * | 2021-06-03 | 2021-09-07 | 安徽理工大学 | Countermeasure sample generation method based on generation model |
| CN113395280A (en) * | 2021-06-11 | 2021-09-14 | 成都为辰信息科技有限公司 | Anti-confusion network intrusion detection method based on generation of countermeasure network |
| CN113505886A (en) * | 2021-07-08 | 2021-10-15 | 深圳市网联安瑞网络科技有限公司 | Countermeasure sample generation method, system, terminal and medium based on fuzzy test |
| CN113537381A (en) * | 2021-07-29 | 2021-10-22 | 大连海事大学 | Human body rehabilitation exercise data enhancement method based on confrontation sample |
| CN113642772A (en) * | 2021-07-13 | 2021-11-12 | 重庆科技学院 | Logging reservoir identification and prediction method based on machine learning |
| CN114548300A (en) * | 2019-12-20 | 2022-05-27 | 支付宝(杭州)信息技术有限公司 | Method and device for explaining service processing result of service processing model |
| WO2022116743A1 (en) * | 2020-12-03 | 2022-06-09 | International Business Machines Corporation | Generating data based on pre-trained models using generative adversarial models |
| CN114663946A (en) * | 2022-03-21 | 2022-06-24 | 中国电信股份有限公司 | Countermeasure sample generation method, apparatus, device and medium |
-
2019
- 2019-05-29 CN CN201910459852.XA patent/CN110334806A/en active Pending
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110768971A (en) * | 2019-10-16 | 2020-02-07 | 伍军 | Confrontation sample rapid early warning method and system suitable for artificial intelligence system |
| CN112884143A (en) * | 2019-11-29 | 2021-06-01 | 北京四维图新科技股份有限公司 | Method for training robust deep neural network model |
| CN112884143B (en) * | 2019-11-29 | 2024-05-14 | 北京四维图新科技股份有限公司 | Method for training robust deep neural network model |
| CN114548300B (en) * | 2019-12-20 | 2024-05-28 | 支付宝(杭州)信息技术有限公司 | Method and device for explaining service processing result of service processing model |
| CN114548300A (en) * | 2019-12-20 | 2022-05-27 | 支付宝(杭州)信息技术有限公司 | Method and device for explaining service processing result of service processing model |
| CN111160217A (en) * | 2019-12-25 | 2020-05-15 | 中山大学 | Method and system for generating confrontation sample of pedestrian re-identification system |
| CN111160217B (en) * | 2019-12-25 | 2023-06-23 | 中山大学 | Method and system for generating countermeasure sample of pedestrian re-recognition system |
| CN111163472A (en) * | 2019-12-30 | 2020-05-15 | 浙江工业大学 | Signal identification attack defense method based on generative countermeasure network |
| CN111210002A (en) * | 2019-12-30 | 2020-05-29 | 北京航空航天大学 | Multi-layer academic network community discovery method and system based on generation of confrontation network model |
| CN111210002B (en) * | 2019-12-30 | 2022-01-28 | 北京航空航天大学 | Multi-layer academic network community discovery method and system based on generation of confrontation network model |
| CN111163472B (en) * | 2019-12-30 | 2022-10-04 | 浙江工业大学 | Signal identification attack defense method based on generative countermeasure network |
| CN111241287A (en) * | 2020-01-16 | 2020-06-05 | 支付宝(杭州)信息技术有限公司 | Training method and device for generating generation model of confrontation text |
| CN111310802A (en) * | 2020-01-20 | 2020-06-19 | 星汉智能科技股份有限公司 | Anti-attack defense training method based on generation of anti-network |
| CN111275115A (en) * | 2020-01-20 | 2020-06-12 | 星汉智能科技股份有限公司 | Method for generating counterattack sample based on generation counternetwork |
| CN111340066A (en) * | 2020-02-10 | 2020-06-26 | 电子科技大学 | Confrontation sample generation method based on geometric vector |
| CN111340066B (en) * | 2020-02-10 | 2022-05-31 | 电子科技大学 | Confrontation sample generation method based on geometric vector |
| CN111539184A (en) * | 2020-04-29 | 2020-08-14 | 上海眼控科技股份有限公司 | Text data manufacturing method and device based on deep learning, terminal and storage medium |
| CN111898645A (en) * | 2020-07-03 | 2020-11-06 | 贵州大学 | Movable sample attack resisting method based on attention mechanism |
| CN111967592A (en) * | 2020-07-09 | 2020-11-20 | 中国电子科技集团公司第三十六研究所 | Method for generating counterimage machine recognition based on positive and negative disturbance separation |
| CN111967592B (en) * | 2020-07-09 | 2023-12-05 | 中国电子科技集团公司第三十六研究所 | Method for generating countermeasure image machine identification based on separation of positive and negative disturbance |
| CN111967584A (en) * | 2020-08-19 | 2020-11-20 | 北京字节跳动网络技术有限公司 | Method, device, electronic equipment and computer storage medium for generating countermeasure sample |
| CN111738374B (en) * | 2020-08-28 | 2020-11-24 | 北京智源人工智能研究院 | Multi-sample anti-disturbance generation method and device, storage medium and computing equipment |
| CN111738374A (en) * | 2020-08-28 | 2020-10-02 | 北京智源人工智能研究院 | Multi-sample anti-disturbance generation method and device, storage medium and computing equipment |
| CN111818101B (en) * | 2020-09-09 | 2020-12-11 | 平安国际智慧城市科技股份有限公司 | Network security detection method and device, computer equipment and storage medium |
| CN111818101A (en) * | 2020-09-09 | 2020-10-23 | 平安国际智慧城市科技股份有限公司 | Network security detection method and device, computer equipment and storage medium |
| CN112162515B (en) * | 2020-10-10 | 2021-08-03 | 浙江大学 | Anti-attack method for process monitoring system |
| CN112162515A (en) * | 2020-10-10 | 2021-01-01 | 浙江大学 | Anti-attack method for process monitoring system |
| US20220180203A1 (en) * | 2020-12-03 | 2022-06-09 | International Business Machines Corporation | Generating data based on pre-trained models using generative adversarial models |
| GB2617722A (en) * | 2020-12-03 | 2023-10-18 | Ibm | Generating data based on pre-trained models using generative adversarial models |
| WO2022116743A1 (en) * | 2020-12-03 | 2022-06-09 | International Business Machines Corporation | Generating data based on pre-trained models using generative adversarial models |
| CN112818407A (en) * | 2021-04-16 | 2021-05-18 | 中国工程物理研究院计算机应用研究所 | Video privacy protection method based on generation countermeasure network |
| CN112818407B (en) * | 2021-04-16 | 2021-06-22 | 中国工程物理研究院计算机应用研究所 | Video privacy protection method based on generation countermeasure network |
| CN113158190A (en) * | 2021-04-30 | 2021-07-23 | 河北师范大学 | Malicious code countermeasure sample automatic generation method based on generation type countermeasure network |
| CN113177599B (en) * | 2021-05-10 | 2023-11-21 | 南京信息工程大学 | Reinforced sample generation method based on GAN |
| CN113177599A (en) * | 2021-05-10 | 2021-07-27 | 南京信息工程大学 | Enhanced sample generation method based on GAN |
| CN113361594A (en) * | 2021-06-03 | 2021-09-07 | 安徽理工大学 | Countermeasure sample generation method based on generation model |
| CN113361594B (en) * | 2021-06-03 | 2023-10-20 | 安徽理工大学 | Countermeasure sample generation method based on generation model |
| CN113395280B (en) * | 2021-06-11 | 2022-07-26 | 成都为辰信息科技有限公司 | Anti-confusion network intrusion detection method based on generation countermeasure network |
| CN113222480B (en) * | 2021-06-11 | 2023-05-12 | 支付宝(杭州)信息技术有限公司 | Training method and device for challenge sample generation model |
| CN113222480A (en) * | 2021-06-11 | 2021-08-06 | 支付宝(杭州)信息技术有限公司 | Training method and device for confrontation sample generation model |
| CN113395280A (en) * | 2021-06-11 | 2021-09-14 | 成都为辰信息科技有限公司 | Anti-confusion network intrusion detection method based on generation of countermeasure network |
| CN113505886A (en) * | 2021-07-08 | 2021-10-15 | 深圳市网联安瑞网络科技有限公司 | Countermeasure sample generation method, system, terminal and medium based on fuzzy test |
| CN113642772A (en) * | 2021-07-13 | 2021-11-12 | 重庆科技学院 | Logging reservoir identification and prediction method based on machine learning |
| CN113537381B (en) * | 2021-07-29 | 2024-05-10 | 大连海事大学 | Human rehabilitation exercise data enhancement method based on countermeasure sample |
| CN113537381A (en) * | 2021-07-29 | 2021-10-22 | 大连海事大学 | Human body rehabilitation exercise data enhancement method based on confrontation sample |
| CN114663946A (en) * | 2022-03-21 | 2022-06-24 | 中国电信股份有限公司 | Countermeasure sample generation method, apparatus, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110334806A (en) | A kind of confrontation sample generating method based on production confrontation network | |
| Rozsa et al. | Are accuracy and robustness correlated | |
| Rozsa et al. | LOTS about attacking deep features | |
| Cao et al. | Adversarial learning with local coordinate coding | |
| Liu et al. | Transductive centroid projection for semi-supervised large-scale recognition | |
| CN112446423A (en) | Fast hybrid high-order attention domain confrontation network method based on transfer learning | |
| CN111047054A (en) | Two-stage countermeasure knowledge migration-based countermeasure sample defense method | |
| He et al. | Transferable sparse adversarial attack | |
| CN113627543B (en) | Anti-attack detection method | |
| Liu et al. | APSNet: Toward adaptive point sampling for efficient 3D action recognition | |
| Rajani et al. | Stacking with auxiliary features for visual question answering | |
| Che et al. | SMGEA: A new ensemble adversarial attack powered by long-term gradient memories | |
| Wang et al. | Occluded person re-identification via defending against attacks from obstacles | |
| Ding et al. | Beyond universal person re-identification attack | |
| CN111967592A (en) | Method for generating counterimage machine recognition based on positive and negative disturbance separation | |
| Dong et al. | Erasing, transforming, and noising defense network for occluded person re-identification | |
| Fatemifar et al. | Face spoofing detection ensemble via multistage optimisation and pruning | |
| CN113935396A (en) | Manifold theory-based method and related device for resisting sample attack | |
| Lv et al. | Chinese character CAPTCHA recognition based on convolution neural network | |
| Wang et al. | Deep fusion: Crafting transferable adversarial examples and improving robustness of industrial artificial intelligence of things | |
| Wang et al. | Generating semantic adversarial examples via feature manipulation | |
| Kwon et al. | Face friend-safe adversarial example on face recognition system | |
| Rami et al. | Source-guided similarity preservation for online person re-identification | |
| Zhou et al. | Improving robustness of random forest under label noise | |
| CN116051924B (en) | Divide-and-conquer defense method for image countermeasure sample |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191015 |