EAGER: Constraint Aware Generative Adversarial Networks

EAGER：约束感知生成对抗网络

• 批准号: 1841119
• 负责人: Xintao Wu
• 金额: $ 10万
• 依托单位: University of Arkansas
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-09-01 至 2020-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1841119&HistoricalAwards=false
• 关键词: EAGER; Constraint; Aware; Generative; Adversarial

Generative adversarial networks (GAN), which estimate the real data distribution through an adversarial game, have achieved great success in image generation and text generation. However, many decision models from real world applications are trained on relational data that contains both numerical and categorical attributes. Furthermore, real world applications often have legal or ethical requirements on the generated data, e.g., discrimination free or privacy preservation, or expect to generate complementary samples to their existing data that may only contain normal samples. The research significantly improves the applicability of generative adversarial networks from image/text data to relational data with application requirements and contributes to the limited base of knowledge in the area of using generative adversarial networks for constraint aware relational data generation. The findings, tools, software code, and curricular materials documents are disseminated to the research community, IT industry, and users, which expects to help domain users generate constraint aware data to meet their business needs. This EAGER research develops novel techniques that enable the current generative adversarial networks to generate realistic relational data with constraints. The developed framework adds a decoder to the generator to generate both numerical and categorical data and incorporates constraint terms into the objective functions of generator and discriminator or introduces multiple discriminators to enforce requirement constraints. The research adopts f-divergences to analyze the convergence of the constraint aware GAN framework when complex constraints are introduced. The research then focuses on developing under the unifying framework two specific models, fair GAN for generating discrimination-free data, and complementary GAN for generating negative samples when only positive samples are available in the training data. The research conducts empirical evaluations of the framework and two specific models in terms of accuracy and convergence, implements and integrates the developed algorithms into TensorFlow, an open source deep learning software system. The developed framework expects to advance theoretical understanding of generative adversarial networks and the two specific GAN models expect to improve the current research on fairness aware learning and fraud detection. In particular, the fair GAN introduces the new approach of fair data generation based on GAN as current fairness aware learning research mainly adopts data modification. The complimentary GAN outperforms existing one-class classification models for fraud detection by generating complementary samples and enabling the trained discriminator to accurately separate abnormal samples from normal ones.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

生成式对抗网络（GAN）通过对抗博弈来估计真实数据的分布，在图像生成和文本生成方面取得了巨大的成功。然而，来自现实世界应用程序的许多决策模型是在包含数值和分类属性的关系数据上训练的。此外，现实世界的应用通常对生成的数据有法律或道德要求，例如，无歧视或隐私保护，或者期望生成可能只包含正常样本的现有数据的补充样本。该研究显著提高了生成对抗网络从图像/文本数据到具有应用需求的关系数据的适用性，并有助于利用生成对抗网络进行约束感知关系数据生成领域的有限知识基础。这些发现、工具、软件代码和课程材料文档被分发给研究社区、IT行业和用户，它们希望帮助领域用户生成约束感知数据，以满足他们的业务需求。这项EAGER研究开发了新的技术，使当前的生成对抗网络能够生成具有约束的现实关系数据。所开发的框架在生成器上增加了一个解码器来生成数值和分类数据，并在生成器和鉴别器的目标函数中加入约束项或引入多个鉴别器来强制执行需求约束。本研究采用f-散度分析了当引入复杂约束时约束感知GAN框架的收敛性。然后，研究重点是在统一的框架下开发两种特定的模型，即用于生成无歧视数据的公平GAN和用于在训练数据中只有正样本时生成负样本的互补GAN。本研究从准确性和收敛性两个方面对框架和两个具体模型进行了实证评估，并将所开发的算法实现并集成到开源深度学习软件系统TensorFlow中。开发的框架有望推进对生成对抗网络的理论理解，两种特定的GAN模型有望改进当前在公平感知学习和欺诈检测方面的研究。特别是，公平GAN引入了基于GAN的公平数据生成的新方法，目前的公平感知学习研究主要采用数据修改。互补GAN通过生成互补样本，并使训练的鉴别器能够准确地将异常样本与正常样本分开，从而优于现有的单类欺诈检测分类模型。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

One-Class Adversarial Nets for Fraud Detection

• DOI: 10.1609/aaai.v33i01.33011286
• 发表时间: 2018-03
• 期刊: ArXiv
• 作者: Panpan Zheng;Shuhan Yuan;Xintao Wu;Jun Yu Li;Aidong Lu

Achieving Causal Fairness through Generative Adversarial Networks

通过生成对抗网络实现因果公平

• DOI: 10.24963/ijcai.2019/201
• 发表时间: 2019
• 期刊: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
• 作者: Xu, Depeng;Wu, Yongkai;Yuan, Shuhan;Zhang, Lu;Wu, Xintao
• 通讯作者: Wu, Xintao