Adversarial Ranking for Language Generation
This addresses the challenge of synthesizing structured natural language outputs, which is an incremental improvement over existing GANs for language generation tasks.
The paper tackles the problem of generating high-quality language descriptions by proposing RankGAN, a generative adversarial network that ranks collections of sentences rather than classifying them individually, leading to improved language generation as demonstrated on multiple public datasets.
Generative adversarial networks (GANs) have great successes on synthesizing data. However, the existing GANs restrict the discriminator to be a binary classifier, and thus limit their learning capacity for tasks that need to synthesize output with rich structures such as natural language descriptions. In this paper, we propose a novel generative adversarial network, RankGAN, for generating high-quality language descriptions. Rather than training the discriminator to learn and assign absolute binary predicate for individual data sample, the proposed RankGAN is able to analyze and rank a collection of human-written and machine-written sentences by giving a reference group. By viewing a set of data samples collectively and evaluating their quality through relative ranking scores, the discriminator is able to make better assessment which in turn helps to learn a better generator. The proposed RankGAN is optimized through the policy gradient technique. Experimental results on multiple public datasets clearly demonstrate the effectiveness of the proposed approach.