AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Deep Time-Stream Framework for Click-through Rate Prediction by Tracking Interest Evolution

Shu-Ting Shi, Wenhao Zheng, Jun Tang, Qing-Guo Chen, Yao Hu, Jianke Zhu, Ming Li

AAAI 2020 · Probabilistic and scientific modeling › Dynamical systems · approach: Latent dynamics models · 13 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use adaptive time step (ODE) instead of fixed step RNN (ablation)DTS + DNN · Industrial datasetAUC: 0.6532 -> 0.6628Improved
use adaptive time step instead of fixed step (ablation)DTS + DIEN · Industrial datasetAUC: 0.7021 -> 0.7412Improved
use adaptive time step instead of fixed step (ablation)DTS + DIN · Industrial datasetAUC: 0.7002 -> 0.7268Improved
use complex neural-net f instead of simple form (ablation)DTS + DNN · Industrial datasetAUC: 0.6389 -> 0.6628Improved
use complex neural-net f instead of simple form (ablation)DTS + DIEN · Industrial datasetAUC: 0.7045 -> 0.7412Improved
add guide loss (ablation)DTS + DNN · Industrial datasetAUC: 0.6441 -> 0.6628Improved
add guide loss (ablation)DTS + DIEN · Industrial datasetAUC: 0.7116 -> 0.7412Improved
add guide loss (ablation)DTS + DIN · Industrial datasetAUC: 0.7096 -> 0.7268Improved
add guide loss (ablation)DTS + PNN · Industrial datasetAUC: 0.7095 -> 0.6763Hurt
add guide loss (ablation)DTS + Wide&Deep · Industrial datasetAUC: 0.7007 -> 0.701No clear effect
add Time-Stream Module to base CTR modelDNN/PNN/Wide&Deep/DIN/DIEN · Amazon ElectronicAUC: 0.7686 -> 0.7789 (DNN); 0.8453 -> 0.8981 (DIEN)Improved
add Time-Stream Module to base CTR modelDNN/PNN/Wide&Deep/DIN/DIEN · Industrial datasetAUC: 0.6385 -> 0.6628 (DNN); 0.7023 -> 0.7412 (DIEN)Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy