[1609.03126] Energy-based Generative Adversarial Networkcontact arXivarXiv Twitter

We introduce the "Energy-based Generative Adversarial Network" model (EBGAN) which views the discriminator as an energy function that attributes low energies to the regions near the data manifold and higher energies to other regions. Similar to the probabilistic GANs, a generator is seen as being trained to produce contrastive samples with minimal energies, while the discriminator is trained to assign high energies to these generated samples. Viewing the discriminator as an energy function allows to use a wide variety of architectures and loss functionals in addition to the usual binary classifier with logistic output. Among them, we show one instantiation of EBGAN framework as using an auto-encoder architecture, with the energy being the reconstruction error, in place of the discriminator. We show that this form of EBGAN exhibits more stable behavior than regular GANs during training. We also show that a single-scale architecture can be trained to generate high-resolution images.

1 mentions: @shinmura0
Date: 2020/02/13 14:21

Referring Tweets

@shinmura0 最後に追記。このような「デュアルオートエンコーダ」の構造は以下の論文で登場した模様。 ・t.co/V6laNMwrEut.co/waGFpByCwc 私は読んでいない(^^;

Bookmark Comments

Related Entries

Read more Generative Adversarial Networksの基礎と応用について
1 users, 3 mentions 2019/10/24 15:49
Read more StyleGAN: A Style-Based Generator Architecture for Generative Adversarial Networksの実装 - 機械学習奮闘記
5 users, 2 mentions 2019/03/06 11:14
Read more [1811.11264] Synthesizing Tabular Data using Generative Adversarial Networks
1 users, 1 mentions 2019/08/18 11:16
Read more Data Synthesis based on Generative Adversarial Networks (VLDB 2018) 読んだ - 糞糞糞ネット弁慶
0 users, 0 mentions 2019/08/25 10:49
Read more Generative Adversarial Networks @ ICML 2019
5 users, 2 mentions 2019/07/21 11:16