arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.07213cs.LGcs.AIcs.CY
收藏

FARE:具有实用证书的可证明公平表征学习

FARE: Provably Fair Representation Learning with Practical Certificates

ETH Zurich(苏黎世联邦理工学院)

Nikola Jovanović, Mislav Balunović, Dimitar I. Dimitrov, Martin Vechev

AI总结:

针对公平表征学习缺乏可落地公平保证的问题,提出 FARE,通过受限编码器与有限样本高置信上界程序,为任意下游分类器提供紧且实用的公平性证书。

摘要:

公平表征学习(FRL)是一类流行方法,旨在通过数据预处理生成公平分类器。近期监管指令强调需要能够提供实用证书的 FRL 方法,即对在预处理数据上训练的任何下游分类器的不公平性给出可证明上界,从而在实际场景中直接提供保证。创建此类 FRL 方法是一个尚未解决的重要挑战。在本工作中,我们应对该挑战并提出 FARE(Fairness with Restricted Encoders,受限编码器公平性),这是首个具有实用公平性证书的 FRL 方法。FARE 基于我们的关键洞见:限制编码器的表征空间能够推导出实用保证,同时在适当的实例化下仍可实现良好的准确性—公平性权衡,例如我们提出的基于公平树的实例化。为生成实用证书,我们开发并应用了一种统计程序,用于计算在 FARE 嵌入上训练的任何下游分类器不公平性的有限样本高置信上界。在全面实验评估中,我们证明 FARE 生成的实用证书是紧的,并且通常甚至可与先前方法获得的纯经验结果相媲美,这确立了本方法的实用价值。

英文摘要:

Fair representation learning (FRL) is a popular class of methods aiming to produce fair classifiers via data preprocessing. Recent regulatory directives stress the need for FRL methods that provide practical certificates, i.e., provable upper bounds on the unfairness of any downstream classifier trained on preprocessed data, which directly provides assurance in a practical scenario. Creating such FRL methods is an important challenge that remains unsolved. In this work, we address that challenge and introduce FARE (Fairness with Restricted Encoders), the first FRL method with practical fairness certificates. FARE is based on our key insight that restricting the representation space of the encoder enables the derivation of practical guarantees, while still permitting favorable accuracy-fairness tradeoffs for suitable instantiations, such as one we propose based on fair trees. To produce a practical certificate, we develop and apply a statistical procedure that computes a finite sample high-confidence upper bound on the unfairness of any downstream classifier trained on FARE embeddings. In our comprehensive experimental evaluation, we demonstrate that FARE produces practical certificates that are tight and often even comparable with purely empirical results obtained by prior methods, which establishes the practical value of our approach.

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。

↑