课题基金 / 基金详情

FAI: Auditing and Ensuring Fairness in Hard-to-Identify Settings

FAI: Auditing and Ensuring Fairness in Hard-to-Identify Settings
FAI:在难以识别的环境中审核并确保公平性
批准号:
1939704
负责人:
Nathan Kallus
金额:
$38.18万
依托单位:
依托单位国家:
美国
项目类别:
Standard Grant
财政年份:
2020
资助国家:
美国
项目状态:
已结题
起止时间:
2020-05-15 至 2023-04-30

项目摘要

项目成果

Nathan Kallus的其他基金

相似基金

相关文献

中文摘要
翻译
用于高风险决策的人工智能(AI)系统的普及引发了一项紧迫的伦理和法律义务,以避免歧视,并保证受保护阶层的公平。示例应用领域包括信用决策、个性化医疗、有针对性的政策制定以及判刑和保释设置。在这些环境中,公平性通常通过决策对不同群体的不同影响来量化。然而,现有数据的基本限制使审计差异和消除差异变得困难或不可能。例如,在信贷和保险索赔数据中,种族等敏感标签都缺失,而且谁也不知道被预审拘留的被告是否会逃跑。在这两种情况下,这种缺失都导致了无法识别的差异。这些识别问题不仅出现在几乎所有关注公平的应用程序中,它们还打破了大多数现有的公平人工智能方法,并在公平人工智能的理论和实践之间造成了一个紧迫的鸿沟。为了解决这个鸿沟,这个项目开发了一种健壮的理论和方法,用于在很难或不可能确定公平度量的环境中评估和确保公平性。具体地说,该项目将开发:(A)在面临基本识别限制的情况下能够可靠地支持可信结论的公平评估方法;(B)即使公平是不可测量的,也能在设计阶段强有力地执行公平的学习算法。一个关键的方法是认识到识别的局限性,并通过考虑算法在实践中可能引起和可能不引起的差异的可能范围来解决这个问题。该项目确定了几个难以或不可能衡量公平性并需要单独处理的离散环境:未观察到的受保护类成员资格、个性化干预、非二进制算法输出、决策支持算法以及不可忽略的选择性标记。通过解决这些问题,该项目将改变在真实实际环境中对差距的评估,在实际环境中,公平实际上很难在统计上确定,并相应地影响我们如何确保人工智能系统是公平的,并使每个人平等受益。这一影响也将通过与实践者的直接合作直接实现。这一奖项反映了NSF的法定使命,并通过使用基金会的智力优势和更广泛的影响审查标准进行评估,被认为值得支持。
英文摘要
The spread of artificial intelligence (AI) systems for high-stakes decision making gives rise to an urgent ethical and legal imperative to avoid discrimination and guarantee fairness with respect to protected classes. Example application domains include credit decisioning, personalized medicine, targeted policymaking, and sentence and bail setting. In these settings, fairness is often quantified by the decisions' disparate impacts on different groups. However, fundamental limits in the data available make both auditing disparities and eliminating them difficult or impossible. For instance, both in credit and insurance claims data, sensitive labels such as race are missing, and one never knows if a defendant detained pretrial would have fled. In both cases this missingness renders disparities unidentifiable. These identification issues not only arise in nearly every application where fairness is a concern, they also break most existing methods for fair AI and create an urgent gap between the theory and practice of fair AI.Addressing this gap, this project develops a robust theory and methodology for assessing and ensuring fairness in settings where fairness metrics are hard or impossible to pin down. Specifically, the project will develop: (a) fairness assessment methods that can reliably support credible conclusions in the face of fundamental identification limitations; (b) learning algorithms that robustly enforce fairness at the design stage even if fairness is unmeasurable. A key approach is recognizing the limits of identification and addressing it by considering the possible ranges of disparities that an algorithm may and may not induce in practice. The project identifies several discrete settings where fairness is hard or impossible to measure and that require separate treatment: unobserved protected class membership, personalized interventions, non-binary algorithmic outputs, decision-support algorithms, and non-ignorable selective labeling. By tackling these, the project stands to transform the assessment of disparity in real practical settings, where fairness is actually statistically difficult to pin down, and correspondingly impact how we ensure AI systems are fair and benefit everyone equally. This impact will also be directly achieved through direct collaborations with practitioners.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.
期刊论文(23)
专著(0)
科研奖励(0)
会议论文
DOI: 10.48550/arxiv.2206.12081
发表时间: 2022-06
期刊:
影响因子: --
作者: [Masatoshi Uehara;Ayush Sekhari;Jason D. Lee;Nathan Kallus;Wen Sun]
通讯作者: Masatoshi Uehara;Ayush Sekhari;Jason D. Lee;Nathan Kallus;Wen Sun
DOI: --
发表时间: 2023
期刊: Proceedings of Machine Learning Research
影响因子: --
作者: [Andrew Bennett, Nathan Kallus]
通讯作者: Andrew Bennett, Nathan Kallus
DOI: 10.1093/biomet/asad059
发表时间: 2020-06
期刊: ArXiv
影响因子: --
作者: [Nathan Kallus;Masatoshi Uehara]
通讯作者: Nathan Kallus;Masatoshi Uehara
Synthetic Control Analysis of the Short-Term Impact of New York State’s Bail Elimination Act on Aggregate Crime
纽约州保释消除法案对总体犯罪的短期影响的综合控制分析
DOI: 10.1080/2330443x.2023.2267617
发表时间: 2023
期刊: Statistics and Public Policy
影响因子: 1.6
作者: [Zhou, Angela, Koo, Andrew, Kallus, Nathan, Ropac, Rene, Peterson, Richard, Koppel, Stephen, Bergin, Tiffany]
通讯作者: Bergin, Tiffany
共 18 条
    CAREER: Robust Policy Learning for Safe and Reliable Algorithmic Decision Making from Observational Data in Sensitive Applications
    • 批准号:
      1846210
    • 项目类别:
      Continuing Grant
    • 资助金额:
      $50.0万
    • 财政年份:
      2019
    • 负责人:
      Nathan Kallus
    • 依托单位:
    CRII: RI: New Methods for Learning to Personalize from Observational Data with Applications to Precision Medicine and Policymaking
    • 批准号:
      1656996
    • 项目类别:
      Standard Grant
    • 资助金额:
      $17.5万
    • 财政年份:
      2017
    • 负责人:
      Nathan Kallus
    • 依托单位:
    海外基金