Synthetic data generation apparatus based on generative adversarial networks and learning method thereof
A synthetic data generation apparatus according to an embodiment includes a generator for generating synthetic data from an input value, a first discriminator learned to distinguish between actual data and the synthetic data, a second discriminator learned to distinguish between the actual data and the synthetic data while satisfying differential privacy, and a third discriminator learned to distinguish between first synthetic data which is output from the generator learned by the first discriminator and second synthetic data which is output from the generator learned by the second discriminator.
1. A synthetic data generation apparatus based on generative adversarial networks, the apparatus comprising:
a generator configured to generate synthetic data from an input value;
a first discriminator learned to distinguish between actual data and the synthetic data;
a second discriminator learned to distinguish between the actual data and the synthetic data while satisfying differential privacy; and
a third discriminator learned to distinguish between first synthetic data which is output from the generator learned by the first discriminator and second synthetic data which is output from the generator learned by the second discriminator.
2. The apparatus of claim 1 ,
wherein the second discriminator leaned to distinguish between the actual data and the synthetic data while satisfying the differential privacy using any one algorithm of DP-GAN and PATE-GAN.
3. The apparatus of claim 1 ,
wherein the generator is sequentially leaned by the first discriminator, the second discriminator, and the third discriminator.
4. A machine learning method that is performed in a computing device including one or more processors and a memory storing one or more programs executed by the one or more processors, the method comprising:
learning a first discriminator to distinguish between actual data and synthetic data generated from a generator;
primarily learning the generator using the learned first discriminator;
learning a second discriminator to distinguish between the actual data the synthetic data while satisfying differential privacy;
secondarily learning the generator using the learned second discriminator;
learning a third discriminator to distinguish between first synthetic data which is output from the generator learned in the primary learning and second synthetic data which is output from the generator learned in the secondary learning; and
tertiarily learning the generator using the learned third discriminator.
5. The method of claim 4 ,
wherein the learning of the second discriminator performs learning to distinguish between the actual data and the synthetic data while satisfying the differential privacy using any one algorithm of DP-GAN and PATE-GAN.
6. A computing device comprising:
one or more processors;
a memory; and
one or more programs stored in the memory and executed by the one or more processors,
wherein the one or more programs include instructions for performing steps comprising:
learning a first discriminator to distinguish between actual data and synthetic data generated from a generator;
primarily learning the generator using the learned first discriminator;
learning a second discriminator to distinguish between the actual data the synthetic data while satisfying differential privacy;
secondarily learning the generator using the learned second discriminator;
learning a third discriminator to distinguish between first synthetic data which is output from the generator learned in the primary learning and second synthetic data which is output from the generator learned in the secondary learning; and
tertiarily learning the generator using the learned third discriminator.
7. The computing device of claim 6 ,
wherein the learning of the second discriminator performs learning to distinguish between the actual data and the synthetic data while satisfying the differential privacy using any one algorithm of DP-GAN and PATE-GAN.