Auditing Algorithms: Adding Accountability to Automated Authority

这项拨款支持一个研讨会，以协调新兴的研究社区开发“算法审计”，一个研究设计，已显示出诊断算法系统的不良后果的承诺。在金融、媒体、信息、交通或任何计算应用中，基于软件的自动化系统很容易产生设计者无法预见的结果，因此算法审计有可能改善这些系统的设计。在这个意义上，审计的名称来自社会科学的“审计研究”，其中一个功能是在现场实验中操纵。本次研讨会建议将这一新的研究领域结合起来，并编写一份报告，介绍最新技术水平和未来的潜在方向。算法审计的主题将计算机科学，信息科学和社会科学以新颖的组合方式结合在一起。 与会者将有机会阐明他们所面临的挑战，介绍现有的审计方法，并提出研究议程，以提供新的见解，推动科学和造福社会。 这次讲习班讨论的科学问题是阻碍采用和发展信息和通信技术根本进步的主要经济和社会障碍。人类如何看待复杂算法是人机交互和认知科学中的一个基本问题。业界对第三方算法审计很感兴趣，认为这是解决用户对算法系统不满和预防未来问题的一种方式。最近的研究表明，使用“大数据”的算法决策可能会增加不公平和结构性不平等。 政治系统形成了维持和传播社会分层的新途径，并可能助长累积的不利因素。算法系统的不良行为导致了重大的社会挑战，如反竞争行为、欺诈、非法歧视以及经济和社会不平等。

This grant supports a workshop to coordinate the emerging research community developing "algorithm auditing," a research design that has shown promise in diagnosing the unwanted consequences of algorithmic systems. Automated software-based systems in finance, media, information, transportation, or any application of computing can easily create outcomes that are unforeseeable by their designers, so algorithm auditing has the potential to improve the design of these systems. Auditing in this sense takes its name from the social scientific "audit study" where one feature is manipulated in a field experiment. This workshop proposes to coalesce this new area of inquiry and to produce a report characterizing the state of the art and potential future directions. The topic of algorithm auditing brings together computer science, information science, and social science in novel combinations. Participants will have opportunities to articulate challenges that they face, to present existing methods for auditing, and to propose research agendas that can provide new insights that advance science and benefit society. The scientific issues addressed by this workshop are key economic and social impediments to the adoption and development of fundamental advances in information and communication technologies. How humans think about complex algorithms is a fundamental problem in human-computer interaction and cognitive science. There is industrial interest in third-party algorithm auditing as a way to address user dissatisfaction with algorithmic systems and to forestall future problems. Recent research suggests that algorithmic decision-making using "big data" may increase unfairness and structural inequality. Algorithmic systems form a new avenue for the maintenance and transmission of social stratification, and may fuel cumulative disadvantage. Unwanted behavior by algorithmic systems contributes to major societal challenges such as anti-competitive behavior, fraud, unlawful discrimination, and economic and social inequality.