常用评估指标的原理和计算方式

Reading time ~1 minute

记录一次数据下跌的排查

负采样的相关调研

[1] Cross-Batch Negative Sampling for Training Two-Tower Recommenders [2] Negative Sampling for Recommendation [3] DSSMFM: Combining user and item feature interactions for recommendation systems [4] Overview Negative Sampling on Recommendation Systems [5] Negative-Sampling-Paper [6] 推荐系统遇到曝光偏差怎么办?用对比学习! [7] 文献阅读:Mixed Negative Sampling for Learning Two-tower Neural Networks in Recommendations [8] Rethinking InfoNCE: How Many Negative Samples Do You Need? Silde Feature Sample

负采样相关的内容

腾讯广告MVKE实现多目标的用户画像建模

以House Price为例快速构建一个Base Line模型 Continue reading

召回部分代码重构

Published on September 29, 2022

Mind多兴趣召回

Published on September 26, 2022