BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Shaping Recommendations in a Marketplace via User & Content Unders
 tanding - Rishabh Mehrotra (Spotify)
DTSTART:20200221T120000Z
DTEND:20200221T130000Z
UID:TALK136126@talks.cam.ac.uk
CONTACT:Guy Aglionby
DESCRIPTION:Multi-sided marketplaces have witnessed an explosive growth by
  facilitating efficient interactions between multiple stakeholders\, inclu
 ding e.g. buyers and retailers (Amazon)\, guests and hosts (AirBnb)\, ride
 rs and drivers (Uber)\, and listeners and artists (Spotify). A large numbe
 r of such platforms rely on machine learning powered matching engines conn
 ecting consumers with suppliers by acting as a central platform\, thereby 
 finding the right fit and efficiently mediating economic transactions betw
 een the two sides.\n\nIn this talk\, we begin by describing a contextual b
 andit model developed for serving explainable music recommendations to use
 rs and showcase the need for explicitly considering supplier-centric objec
 tives during optimization. We demonstrate how enhanced user and content un
 derstanding helps us in developing better models to power multi-stakeholde
 r marketplaces. Towards the end\, we highlight key NLP challenges faced wh
 en developing such systems to power large scale marketplaces.\n\nBio:\nRis
 habh Mehrotra is a Senior Research Scientist at Spotify Research in London
 . He obtained his PhD in the field of Machine Learning and Information Ret
 rieval from University College London where he was partially supported by 
 a Google Research Award. His current research focuses on marketplace ML an
 d bandit based recommendations. Some of his recent work has been published
  at top conferences including WWW\, SIGIR\, NAACL\, CIKM\, RecSys and WSDM
 . He has co-taught a number of tutorials at leading conferences & multiple
  courses at summer schools.\n
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR