Data and Winning Code for the 2020-2021 SIOP Machine Learning Competition
All important decisions in life involve trade-offs. A potential mate may be stunningly attractive, but what if they are incompatible? You might find the home of your dreams but not in the neighborhood you want. Making great decisions requires balancing competing criteria and finding the optimal outcome. Hiring is no different. To hire effectively one must not only maximize outcomes for the business but also comply with legal requirements. This is often called the “diversity-validity trade-off.” This competition was about developing algorithms that simultaneously maximize business outcomes of job performance and retention while minimizing bias.
The competition portal will provide details about the data, optimization criteria, and FAQs.
Competition Overview and Awards Presentation
Feng Guo @ Bowling Green State University
Sam T. McAbee @ Bowling Green State University
Private Test Set Overall Score = 62.53
Ian Burke @ Axiom
Ashlyn Lowe @ Axiom
Goran Kuljanin @ DePaul University
Robin Burke @ The University of Colorado Boulder
Private Test Set Overall Score = 62.50
Brian Costello @ Red Hat
Willy Hardy @ Red Hat
Private Test Set Overall Score = 61.09
Joshua Prasad @ Colorado State University
Steven Raymer @ Colorado State University
Kelly Cave @ Colorado State University
Shayln Stevens @ Colorado State University
Jason Prasad @ Georgia Institute of Technology
Private Test Set Overall Score = 60.72
Nick Koenig @ Modern Hire
Isaac Thompson @ Modern Hire
Koenig, N., & Thompson, I. The 2020-2021 SIOP Machine Learning Competition. Presented at the 36th annual Society for Industrial and Organizational Psychology conference in New Orleans, LA.