Recently, generative adversarial networks (GANs) have shown promising performance in generating realistic images. However, they often struggle in learning complex underlying modalities in a given dataset, resulting in poor-quality generated images. To mitigate this problem, we present a novel approach called mixture of experts GAN (MEGAN), an ensemble approach of multiple generator networks. Each generator network in MEGAN specializes in generating images with a particular subset of modalities, e.g., an image class. Instead of incorporating a separate step of handcrafted clustering of multiple modalities, our proposed model is trained through an end-to-end learning of multiple generators via gating networks, which is responsible for choosing the appropriate generator network for a given condition. We adopt the categorical reparameterization trick for a categorical decision to be made in selecting a generator while maintaining the flow of the gradients. We demonstrate that individual generators learn different and salient subparts of the data and achieve a multiscale structural similarity (MS-SSIM) score of 0.2470 for CelebA and a competitive unsupervised inception score of 8.33 in CIFAR-10.
|Title of host publication||Proceedings of the 27th International Joint Conference on Artificial Intelligence, IJCAI 2018|
|Publisher||International Joint Conferences on Artificial Intelligence|
|Number of pages||7|
|Publication status||Published - 2018|
|Event||27th International Joint Conference on Artificial Intelligence, IJCAI 2018 - Stockholm, Sweden|
Duration: 2018 Jul 13 → 2018 Jul 19
|Name||IJCAI International Joint Conference on Artificial Intelligence|
|Other||27th International Joint Conference on Artificial Intelligence, IJCAI 2018|
|Period||18/7/13 → 18/7/19|
Bibliographical noteFunding Information:
This work was supported by the National Research Council of Science & Technology (NST) grant by the Korea government (MSIP) [No. CRC-15-05-ETRI].
‡ Corresponding author; This work was supported by the National Research Council of Science & Technology (NST) grant by the Korea government (MSIP) [No. CRC-15-05-ETRI].
All Science Journal Classification (ASJC) codes
- Artificial Intelligence