A Three-Player GAN: Generating Hard Samples To Improve Classification Networks
This addresses robustness in classification networks for applications like traffic sign recognition, but it is incremental as it builds on existing GAN frameworks.
The paper tackles the problem of improving classification network robustness by proposing a Three-Player GAN that generates hard-to-label samples, resulting in a more robust classifier when trained on these samples, as evaluated on a traffic sign recognition dataset.
We propose a Three-Player Generative Adversarial Network to improve classification networks. In addition to the game played between the discriminator and generator, a competition is introduced between the generator and the classifier. The generator's objective is to synthesize samples that are both realistic and hard to label for the classifier. Even though we make no assumptions on the type of augmentations to learn, we find that the model is able to synthesize realistically looking examples that are hard for the classification model. Furthermore, the classifier becomes more robust when trained on these difficult samples. The method is evaluated on a public dataset for traffic sign recognition.