Our paper “Extreme Multi-label Classification for Ad Targeting using Factorization Machines” has been accepted for publication at the KDD 2023 conference Applied Track, and paper “Practical Budget Pacing Algorithms and Simulation Test Bed for eBay Marketplace Sponsored Search” has been accepted at the AdKDD 2023 workshop. Both papers will be presented at the KDD 2023…
Our paper “Predicting Actions of Users using Heterogeneous Online Signals” has been accepted for publication in the Big Data journal!
Our paper “Efficient deep distribution network for bid shading in First-Price Auctions” has been accepted for publication at the KDD 2021 conference!
We shared prospective conversion prediction datasets of time-ordered user activity trails collected across many different sources on the Web. This dataset is specifically designed for prospective advertising, a specific form of targeting where advertisers wish to target only potential and new users. In the updated version of the dataset retargeting signals are included in the additional…
Our papers “Bid Shading in the Brave New World of First Price Auctions” and “Prospective Modeling of Users for Online Display Advertising via Deep Time-Aware Model” have been accepted for publication for the CIKM 2020 conference!