CycleGANAS: Differentiable Neural Architecture Search for CycleGAN
This work provides a novel NAS approach for CycleGAN, potentially benefiting researchers in computer vision and generative models, though it is incremental as it extends existing NAS techniques to a more complex structure.
The authors tackled the problem of unpaired image-to-image translation by developing a Neural Architecture Search (NAS) framework for CycleGAN, which discovers architectures that match or surpass the original CycleGAN's performance and addresses data imbalance through individual searches for each translation direction.
We develop a Neural Architecture Search (NAS) framework for CycleGAN that carries out unpaired image-to-image translation task. Extending previous NAS techniques for Generative Adversarial Networks (GANs) to CycleGAN is not straightforward due to the task difference and greater search space. We design architectures that consist of a stack of simple ResNet-based cells and develop a search method that effectively explore the large search space. We show that our framework, called CycleGANAS, not only effectively discovers high-performance architectures that either match or surpass the performance of the original CycleGAN, but also successfully address the data imbalance by individual architecture search for each translation direction. To our best knowledge, it is the first NAS result for CycleGAN and shed light on NAS for more complex structures.