Internet-advertising-Match & Ranking

计算广告召回&模型&创意算法

1. Match and Targeting (召回/粗排与定向)

• Truncation-Free Matching System for Display Advertising at Alibaba. by Jin Li et al. Arxiv, 2021.
• Learning Optimal Tree Models under Beam Search. by Jingwei Zhuo et al. ICML 2020. (Alibaba Group)
• Sample Optimization For Display Advertising by Hongliang Fei et al. CIKM 2020.
• Deep Retrieval: An End-to-End Learnable Structure Model for Large-Scale Recommendations by Weihao Gao et al. Arxiv, 2020.
• Joint Optimization of Tree-based Index and Deep Model for Recommender Systems. by Han Zhu et al. NeurIPS 2019. (Alibaba Group)
• Learning Tree-based Deep Model for Recommender Systems. by Han Zhu et al. KDD 2018. Code. (Alibaba Group)

中文材料

• 阿里妈妈展示广告粗排：面向链路一致性优化的端到端序学习模型

2. Rank Model

综述材料

• 腾讯技术--一文彻底搞懂CTR建模(1)
• 腾讯技术--一文彻底搞懂CTR建模(2)

延迟反馈建模

• defer:中文链接

论文

长期价值预估

• A DEEP PROBABILISTIC MODEL FOR CUSTOMER LIFETIME VALUE PREDICTION

3. Creatives （创意）

• A Hybrid Bandit Model with Visual Priors for Creative Ranking in Display Advertising. by Shiyao Wang et al. WWW 2021. (Alibaba Group)
• Automated Creative Optimization for E-Commerce Advertising by Jin Chen et al. WWW 2021. (Alibaba Group)
• Efficient Optimal Selection for Composited Advertising Creatives with Tree Structure by Jin Chen et al. AAAI 2021. (Alibaba Group)

4. 特殊Topic

4.1 样本选择偏差问题（Bias）

• An Empirical Study of Selection Bias in Pinterest Ads Retrieval by KDD’23
  • 中文：Pinterest搜索广告: 召回阶段样本选择偏差问题实证分析

4.2 Uplift Modeling

• 一文读懂增益模型Uplift Model

4.3 不确定性建模

• 基于深度置信度感知学习的广告投放探索方案
• 不确定度建模（上）-贝叶斯推断基础