开始:VCALENDAR版本:2.0 PRODID: / /学院Labor Economics//Zope//EN METHOD:PUBLISH CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Europe/Berlin BEGIN:DAYLIGHT TZOFFSETFROM:+0100 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU DTSTART:19810329T020000 TZNAME:CEST TZOFFSETTO:+0200 END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:+0200 RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU DTSTART:19961027T030000 TZNAME:CET TZOFFSETTO:+0100 END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:34061621288800@conference.iza.org LOCATION;CHARSET=UTF-8: DESCRIPTION:This paper views hiring as a contextual bandit problem: to find the best workers over time, firms must balance "exploitation" (selecting from groups with proven track records) with "exploration" (selecting from under-represented groups to learn about quality). Yet modern \nhiring algorithms, based on \supervised learning" approaches, are designed solely for exploitation. \nInstead, we build a resume screening algorithm that values exploration by evaluating candidates \naccording to their statistical upside potential. Using data from professional services recruiting \nwithin a Fortune 500 firm, we show that this approach improves the quality (as measured by \neventual hiring rates) of candidates selected for an interview, while also increasing demographic \ndiversity, relative to the firm's existing practices. The same is not true for traditional supervised \nlearning based algorithms, which improve hiring rates but select far fewer Black and Hispanic \napplicants. In an extension, we show that exploration-based algorithms are also able to learn \nmore effectively about simulated changes in applicant hiring potential over time. Together, our \nresults highlight the importance of incorporating exploration in developing decision-making \nalgorithms that are potentially both more efficient and equitable. SEQUENCE:1 X-APPLE-TRAVEL-ADVISORY-BEHAVIOR:AUTOMATIC SUMMARY:IZA Seminar: Hiring as Exploration by Danielle Li (MIT Sloan School of Management) DTSTART;TZID=Europe/Berlin:20210518T140000 DTEND;TZID=Europe/Berlin:20210518T151500 END:VEVENT END:VCALENDAR