Researchers investigate whether ‘fairness constraints’ mitigate bias in algorithms
Automated systems trained with real-world data have a tendency to discriminate against disadvantaged groups. For instance, an algorithm adopted by the U.K. government downgraded hundreds of thousands of students’ grades, disproportionately impacting those from tuition-free schools. One approach to alleviating the bias issue is imposing “fairness constraints” such that certain statistical measures are roughly equalized across groups. But the long-term […]
Continue reading »