arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.07754cs.LG
收藏

生成模型中的公平性度量

Measuring Fairness in Generative Models

Singapore University of Technology and Design(新加坡科技设计大学)

Christopher T. H Teo, Ngai-Man Cheung

摘要:

深度生成模型在提高训练稳定性和生成数据质量方面取得了很大进展。最近,人们对深层生成数据的公平性越来越感兴趣。公平在许多应用中很重要,例如执法,因为偏见会影响效率。公平数据生成的核心是用于评估和评价不同生成模型的公平性度量。在本文中,我们首先回顾公平性指标提出了在以前的作品,并突出潜在的弱点。然后,我们讨论了性能基准框架以及替代指标的评估。

英文摘要:

Deep generative models have made much progress in improving training stability and quality of generated data. Recently there has been increased interest in the fairness of deep-generated data. Fairness is important in many applications, e.g. law enforcement, as biases will affect efficacy. Central to fair data generation are the fairness metrics for the assessment and evaluation of different generative models. In this paper, we first review fairness metrics proposed in previous works and highlight potential weaknesses. We then discuss a performance benchmark framework along with the assessment of alternative metrics.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑