FSCD

参考资料
[1] Towards a Better Tradeoff between Effectiveness and Efficiency in Pre-Ranking: A Learnable Feature Selection based Approach. SIGIR 2021 （知乎解读）
[2] Learning Sparse Neural Networks through L0 Regularization. ICLR 2018
[3] Dropout rank : https://arxiv.org/pdf/1712.08645.pdf

LPFS

论文：LPFS: Learnable Polarizing Feature Selection for Click-Through Rate Prediction
链接：https://arxiv.org/pdf/2206.00267

加压删特征