DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jeon, Insu | ko |
dc.contributor.author | Lee, Wonkwang | ko |
dc.contributor.author | Pyeon, Myeongjang | ko |
dc.contributor.author | Kim, Gunhee | ko |
dc.date.accessioned | 2021-11-01T06:42:06Z | - |
dc.date.available | 2021-11-01T06:42:06Z | - |
dc.date.created | 2021-10-27 | - |
dc.date.created | 2021-10-27 | - |
dc.date.issued | 2021-02 | - |
dc.identifier.citation | 35th AAAI Conference on Artificial Intelligence / 33rd Conference on Innovative Applications of Artificial Intelligence / 11th Symposium on Educational Advances in Artificial Intelligence, pp.7926 - 7934 | - |
dc.identifier.issn | 2159-5399 | - |
dc.identifier.uri | http://hdl.handle.net/10203/288488 | - |
dc.description.abstract | We propose a new GAN-based unsupervised model for disentangled representation learning. The new model is discovered in an attempt to utilize the Information Bottleneck (IB) framework to the optimization of GAN, thereby named IB-GAN. The architecture of IB-GAN is partially similar to that of InfoGAN but has a critical difference; an intermediate layer of the generator is leveraged to constrain the mutual information between the input and the generated output. The intermediate stochastic layer can serve as a learnable latent distribution that is trained with the generator jointly in an end-to-end fashion. As a result, the generator of IB-GAN can harness the latent space in a disentangled and interpretable manner. With the experiments on dSprites and Color-dSprites dataset, we demonstrate that IB-GAN achieves competitive disentanglement scores to those of state-of-the-art beta-VAEs and outperforms InfoGAN. Moreover, the visual quality and the diversity of samples generated by IB-GAN are often better than those by beta-VAEs and Info-GAN in terms of FID score on CelebA and 3D Chairs dataset. | - |
dc.language | English | - |
dc.publisher | ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE | - |
dc.title | IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks | - |
dc.type | Conference | - |
dc.identifier.wosid | 000680423508005 | - |
dc.identifier.scopusid | 2-s2.0-85121306656 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 7926 | - |
dc.citation.endingpage | 7934 | - |
dc.citation.publicationname | 35th AAAI Conference on Artificial Intelligence / 33rd Conference on Innovative Applications of Artificial Intelligence / 11th Symposium on Educational Advances in Artificial Intelligence | - |
dc.identifier.conferencecountry | CN | - |
dc.identifier.conferencelocation | ELECTR NETWORK | - |
dc.contributor.localauthor | Lee, Wonkwang | - |
dc.contributor.nonIdAuthor | Jeon, Insu | - |
dc.contributor.nonIdAuthor | Pyeon, Myeongjang | - |
dc.contributor.nonIdAuthor | Kim, Gunhee | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.