课题基金 / 基金详情

A software solution to tackle algorithmic bias

A software solution to tackle algorithmic bias
解决算法偏差的软件解决方案
批准号:
53630
负责人:
金额:
$47.0万
依托单位:
依托单位国家:
英国
项目类别:
Study
财政年份:
2020
资助国家:
英国
项目状态:
已结题
起止时间:
2020 至 --

项目摘要

项目成果

相似基金

相关文献

中文摘要
翻译
ETIQ是一种软件解决方案,可以帮助公司识别和减轻自动化决策中的偏见。无论是在警察和判刑、招聘还是金融产品方面,人们的生活每天都受到这些自动化决策的影响。我们是机器学习等先进技术的坚定倡导者,并坚信如果正确使用该技术,它将为每个人提供更公平的过程和更平等的机会。我们的目标是确保我们使用的算法产生的决策与我们的预期一致。当我们发现算法偏差的证据时,我们的目标是调整这些算法,以获得更公平的结果。金融公司的目标是开始使用更丰富的数据和更复杂的算法,以进一步优化其自动决策,从而提高它们的利润。然而,在缺乏偏差评估和偏差缓解计划和解决方案的情况下,由于监管,他们无法将其改进的算法投入生产,希望成为可持续和符合道德的。此外,使用现成的偏差识别器和缓解解决方案意味着更快的上市时间和更短的算法部署测试时间。
英文摘要
ETIQ is a software solution that helps companies identify and mitigate bias in their automated decisions. Whether it's in policing and sentencing, recruitment or financial products, people's lives are being affected by these automated decisions on a daily basis. We are strong advocates of advanced technology such as machine learning and firmly believe that if the technology is used correctly, it will provide a fairer process and more equal opportunities to everyone. Our goal is to ensure that the algorithms we use produce decisions that are consistent with our expectations. When we find evidence of algorithmic bias, we aim to adjust these algorithms to obtain fairer results.Financial companies aim to start using richer data and more sophisticated algorithms to optimise their automated decision making further and thus improve their bottom line. However in the lack of a bias assessment and bias mitigation plan and solution they are unable to put into production their improved algorithms due to regulation, desire to be sustainable and ethical. Furthermore, using a bias identifier and mitigation solution that's off the shelf means faster time to market and shorter testing times for an algorithm deployment.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
国内基金
海外基金
Navigating Sustainability: Understanding Environm ent,Social and Governanc e Challenges and Solution s for Chinese Enterprises in Pakistan's CPEC Framew ork
  • 批准号:
    --
  • 项目类别:
    外国学者研究基金项目
  • 资助金额:
    --
  • 批准年份:
    2024
  • 负责人:
    Noshaba Aziz
  • 依托单位:
Horndeski理论中Randall-Sundrum型厚膜解的研究
  • 批准号:
    11605127
  • 项目类别:
    青年科学基金项目
  • 资助金额:
    18.0万元
  • 批准年份:
    2016
  • 负责人:
    钟渊
  • 依托单位:
N-体问题的中心构型及动力系统的分支理论
  • 批准号:
    10601071
  • 项目类别:
    青年科学基金项目
  • 资助金额:
    10.0万元
  • 批准年份:
    2006
  • 负责人:
    朱长荣
  • 依托单位: