로그인
통찰 - Uplift Modeling with Graph Neural Networks
Leveraging Graph Neural Networks to Predict Treatment Effects with Limited Supervision
The core message of this paper is to develop a novel modular framework, based on graph neural networks (GNNs) and active learning, to address the need for limited supervision in uplift modeling, moving from the standard "70%-80%" train set rule down to 5%-20%.
Graph Neural Network with Two Uplift Estimators for Label-Scarcity Individual Uplift Modeling
The author proposes a novel framework, GNUM, utilizing graph neural networks and two uplift estimators to address label scarcity in individual uplift modeling.