Deep neural networks

Alimama Tech

May 25, 2022 · Artificial Intelligence

AdaCalib: Posterior-Guided Feature-Adaptive Calibration Model for Online Advertising

AdaCalib is a posterior‑guided feature‑adaptive calibration model for online advertising that learns per‑feature piecewise‑linear calibration functions via a deep neural network with adaptive bucketing, improving probability estimates and ranking, achieving lower Field‑RCE, higher AUC, and a 5% CVR lift in live tests.

Deep neural networkscalibrationfeature adaptation

0 likes · 19 min read

AdaCalib: Posterior-Guided Feature-Adaptive Calibration Model for Online Advertising

Baidu Geek Talk

Aug 16, 2021 · Artificial Intelligence

End-to-End Consistency Testing Solution for Click-Through Rate Models in Advertising Systems

The article describes Baidu’s end-to-end consistency testing framework for advertising click-through-rate models, which uses a five-stream verification pipeline and six implementation phases to compare Q-values across feature extraction, table conversions, and DNN computation, enabling precise detection and localization of data and model inconsistencies in production.

Advertising SystemsBaiduCTR prediction

0 likes · 17 min read

End-to-End Consistency Testing Solution for Click-Through Rate Models in Advertising Systems

iQIYI Technical Product Team

Nov 10, 2017 · Artificial Intelligence

iQIYI Recommendation System: Architecture, Model Evolution, and Ranking Strategies

The iQIYI recommendation system combines a two‑stage pipeline of recall and ranking, evolving from logistic regression to a GBDT‑FM‑DNN ensemble, using online feature storage, extensive feature engineering, and configurable strategies to deliver personalized video suggestions while addressing feature drift and multi‑objective business goals.

Deep neural networksFactorization MachinesGBDT

0 likes · 13 min read

iQIYI Recommendation System: Architecture, Model Evolution, and Ranking Strategies