Maximum-Likelihood Augmented Discrete Generative Adversarial Networks

02/26/2017
by   Tong Che, et al.
0

Despite the successes in capturing continuous distributions, the application of generative adversarial networks (GANs) to discrete settings, like natural language tasks, is rather restricted. The fundamental reason is the difficulty of back-propagation through discrete random variables combined with the inherent instability of the GAN training objective. To address these problems, we propose Maximum-Likelihood Augmented Discrete Generative Adversarial Networks. Instead of directly optimizing the GAN objective, we derive a novel and low-variance objective using the discriminator's output that follows corresponds to the log-likelihood. Compared with the original, the new objective is proved to be consistent in theory and beneficial in practice. The experimental results on various discrete datasets demonstrate the effectiveness of the proposed approach.

READ FULL TEXT
02/27/2017

Boundary-Seeking Generative Adversarial Networks

We introduce a novel approach to training generative adversarial network...
09/05/2019

Learning from Label Proportions with Generative Adversarial Networks

In this paper, we leverage generative adversarial networks (GANs) to der...
01/28/2022

Generative Cooperative Networks for Natural Language Generation

Generative Adversarial Networks (GANs) have known a tremendous success f...
04/04/2019

Generative Adversarial Networks for text using word2vec intermediaries

Generative adversarial networks (GANs) have shown considerable success, ...
03/14/2019

Deep Switch Networks for Generating Discrete Data and Language

Multilayer switch networks are proposed as artificial generators of high...
07/09/2021

InfoVAEGAN : learning joint interpretable representations by information maximization and maximum likelihood

Learning disentangled and interpretable representations is an important ...
08/04/2022

A Representation Modeling Based Language GAN with Completely Random Initialization

Text generative models trained via Maximum Likelihood Estimation (MLE) s...

Please sign up or login with your details

Forgot password? Click here to reset