{"id":68498,"date":"2019-10-17T10:17:55","date_gmt":"2019-10-17T14:17:55","guid":{"rendered":"http:\/\/danielschristian.com\/learning-ecosystems\/?p=68498"},"modified":"2019-10-17T11:17:49","modified_gmt":"2019-10-17T15:17:49","slug":"can-you-make-ai-fairer-than-a-judgeplus-several-other-legal-related-items","status":"publish","type":"post","link":"http:\/\/danielschristian.com\/learning-ecosystems\/2019\/10\/17\/can-you-make-ai-fairer-than-a-judgeplus-several-other-legal-related-items\/","title":{"rendered":"“Can you make AI fairer than a judge? Play our courtroom algorithm game” [Hao & Stray] Plus several other legal-related items."},"content":{"rendered":"

Can you make AI fairer than a judge? Play our courtroom algorithm game<\/strong><\/a> — from technologyreview.com by Karen Hao and Jonathan Stray
\nPlay our courtroom algorithm game The US criminal legal system uses predictive algorithms to try to make the judicial process less biased. But there\u2019s a deeper problem.<\/em><\/p>\n

Excerpt:<\/em><\/p>\n

As a child, you develop a sense of what \u201cfairness\u201d means. It\u2019s a concept that you learn early on as you come to terms with the world around you. Something either feels fair or it doesn\u2019t.<\/span><\/p>\n

But increasingly, algorithms have begun to arbitrate fairness for us. They decide who sees housing ads, who gets hired or fired, and even who gets sent to jail. Consequently, the people who create them\u2014software engineers\u2014are being asked to articulate what it means to be fair in their code. This is why regulators around the world are now grappling with a question: How can you mathematically quantify fairness?\u00a0<\/span><\/p>\n

This story attempts to offer an answer. And to do so, we need your help. We\u2019re going to walk through a real algorithm, one used to decide who gets sent to jail, and ask you to tweak its various parameters to make its outcomes more fair. (Don\u2019t worry\u2014this won\u2019t involve looking at code!)<\/span><\/p>\n

The algorithm we\u2019re examining<\/strong>\u00a0is known as COMPAS, and it\u2019s one of several different \u201crisk assessment\u201d tools used in the US criminal legal system.<\/p>\n

 <\/p>\n

But whether algorithms should be used to arbitrate fairness in the first place is a complicated question. Machine-learning algorithms are trained on \u201cdata produced through histories of exclusion and discrimination,\u201d writes Ruha Benjamin, an associate professor at Princeton University, in her book\u00a0Race After Technology<\/em>. Risk assessment tools are no different. The greater question about using them\u2014or any algorithms used to rank people\u2014is whether they reduce existing inequities or make them worse.<\/strong><\/span><\/p><\/blockquote>\n

 <\/p>\n

You can also see change in these articles as well:<\/em><\/p>\n