Ad click prediction

H. Brendan McMahan(Google (United States)), Gary D. Holt(Google (United States)), D. Sculley(Google (United States)), Michael Young(Google (United States)), Dietmar Ebner(Google (United States)), Julian Grady(Google (United States)), Lan Nie(Google (United States)), Todd Phillips(Google (United States)), Eugene Davydov(Google (United States)), Daniel Golovin(Google (United States)), Sharat Chikkerur(Google (United States)), Dan Liu(Google (United States)), Martin Wattenberg(Google (United States)), Arnar Mar Hrafnkelsson(Google (United States)), Tom Boulos(Google (United States)), Jeremy Kubica(Google (United States))
Unknown
August 11, 2013
Cited by 880Open Access
Full Text

Abstract

Predicting ad click-through rates (CTR) is a massive-scale learning problem that is central to the multi-billion dollar online advertising industry. We present a selection of case studies and topics drawn from recent experiments in the setting of a deployed CTR prediction system. These include improvements in the context of traditional supervised learning based on an FTRL-Proximal online learning algorithm (which has excellent sparsity and convergence properties) and the use of per-coordinate learning rates.


Related Papers

No related papers found

Powered by citation graph analysis