Adversarial Ranking for Language Generation

TL;DR

RankGAN introduces a ranking-based discriminator in GANs to improve language generation by evaluating sentence quality relative to a group, leading to higher-quality outputs.

Contribution

It proposes a novel ranking-based discriminator for GANs, enhancing language generation by assessing sentence quality through relative ranking rather than binary classification.

Findings

RankGAN outperforms traditional GANs on multiple datasets.

The ranking approach improves the quality of generated language descriptions.

Experimental results validate the effectiveness of the proposed method.

Abstract

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.