Method for training and testing data embedding network to generate marked data by integrating original data with mark data, and training device and testing device using the same

A method for testing a learned data embedding network capable of integrating original data with mark data to thereby generate marked data, comprising steps of:

• (a) on condition that a learning device has performed processes of (i) if original training data and mark training data have been acquired, inputting the original training data and the mark training data into a data embedding network, and allowing the data embedding network to integrate the original training data and the mark training data and thus to generate marked training data, (ii) inputting the marked training data into a learning network having its own learned parameters, and allowing the learning network to apply a network operation to the marked training data using the learned parameters and thus to generate 1-st characteristic information corresponding to the marked training data, and inputting the original training data into the learning network, and allowing the learning network to apply a network operation to the original training data using the learned parameters and thus to output 2-nd characteristic information corresponding to the original training data, (iii) learning the data embedding network, such that at least one data error is minimized which is calculated by referring to at least part of (iii-1) at least one 1-st error acquired by referring to the 1-st characteristic information and the 2-st characteristic information and (iii-2) at least one 2-nd error acquired by referring to at least one task specific output generated by using the 1-st characteristic information and by further referring to at least one ground truth corresponding to the task specific output and such that at least one marked data score is maximized which corresponds to the marked training data inputted into a discriminator for determining whether inputted data is real or fake, and (iv) learning the discriminator such that at least one modified data score or at least one modified marked data score is maximized which respectively corresponds to modified training data or modified marked training data inputted into the discriminator and such that the marked data score is minimized wherein the modified training data or the modified marked training data is generated respectively by modifying the original training data or the marked training data, a testing device performing a process of acquiring original test data and mark test data to be integrated;
• and
• (b) the testing device inputting the original test data and the mark test data into the data embedding network, and allowing the data embedding network to integrate the original test data with the mark test data using one or more learned parameters of the data embedding network and thus to generate marked test data.

A learning device for learning a data embedding network capable of integrating original data with mark data to thereby generate marked data, comprising:

• at least one memory that stores instructions;
• and
• at least one processor configured to execute the instructions to perform or support another device to perform processes of:
• (I) if original training data and mark training data are acquired, inputting the original training data and the mark training data into the data embedding network, and allowing the data embedding network to integrate the original training data with the mark training data and thus to generate marked training data, (II) (i) inputting the marked training data into a learning network having its own one or more learned parameters, and allowing the learning network to (i-1) apply a network operation to the marked training data using the learned parameters and thus to (i-2) generate 1-st characteristic information corresponding to the marked training data, and (ii) inputting the original training data into the learning network, and allowing the learning network to (ii-1) apply a network operation to the original training data using the learned parameters and thus to (ii-2) generate 2-nd characteristic information corresponding to the original training data, and (III) (III-1) learning the data embedding network such that (i) at least one data error is minimized which is calculated by referring to at least part of (i-1) at least one 1-st error acquired by referring to the 1-st characteristic information and the 2-nd characteristic information, and (i-2) at least one 2-nd error acquired by referring to at least one task specific output generated by using the 1-st characteristic information and by further referring to at least one ground truth corresponding to the task specific output, and such that (ii) at least one marked data score is maximized which corresponds to the marked training data inputted into a discriminator for determining whether inputted data is real or fake, and (III-2) learning the discriminator such that at least one modified data score or at least one modified marked data score is maximized which respectively corresponds to modified training data or modified marked training data inputted into the discriminator and such that the marked data score is minimized wherein the modified training data or the modified marked training data is generated respectively by modifying the original training data or the marked training data.

A testing device for testing a learned data embedding network capable of integrating original data with mark data to thereby generate marked data, comprising:

• at least one memory that stores instructions;
• and
• at least one processor configured to execute the instructions to perform or support another device to perform processes of:
• on condition that a learning device has performed processes of (i) if original training data and mark training data have been acquired, inputting the original training data and the mark training data into a data embedding network, and allowing the data embedding network to integrate the original training data and the mark training data and thus to generate marked training data, (ii) inputting the marked training data into a learning network having its own learned parameters, and allowing the learning network to apply a network operation to the marked training data using the learned parameters and thus to generate 1-st characteristic information corresponding to the marked training data, and inputting the original training data into the learning network, and allowing the learning network to apply a network operation to the original training data using the learned parameters and thus to output 2-nd characteristic information corresponding to the original training data, (iii) learning the data embedding network, such that at least one data error is minimized which is calculated by referring to at least part of (iii-1) at least one 1-st error acquired by referring to the 1-st characteristic information and the 2-st characteristic information and (iii-2) at least one 2-nd error acquired by referring to at least one task specific output generated by using the 1-st characteristic information and by further referring to at least one ground truth corresponding to the task specific output and such that at least one marked data score is maximized which corresponds to the marked training data inputted into a discriminator for determining whether inputted data is real or fake, and (iv) learning the discriminator such that at least one modified data score or at least one modified marked data score is maximized which corresponds to modified training data or modified marked training data inputted into the discriminator and such that the marked data score is minimized wherein the modified training data or the modified marked training data is generated respectively by modifying the original training data or the marked training data, processes of (I) acquiring original test data and mark test data to be integrated;
• and (II) inputting the original test data and the mark test data into the data embedding network, and allowing the data embedding network to integrate the original test data with the mark test data using one or more learned parameters of the data embedding network and thus to generate marked test data.