推荐benreesman · 2024-03
“point the interested at the DLRM paper [1]”
向感兴趣的人推荐 DLRM 论文
Meta 开源的深度学习推荐模型 DLRM 及其并行训练方案
论文提出深度学习推荐模型 DLRM,并给出 PyTorch 和 Caffe2 两种实现。针对嵌入表带来的内存压力,设计了对嵌入表做模型并行、对全连接层做数据并行的并行方案。作者在 Big Basin 平台上对其性能做了刻画,并将 DLRM 定位为未来算法实验和系统协同设计的基准。
评论者把它当作大型推荐网络的参考起点,常用它说明嵌入查找是性能和内存瓶颈,也有人提醒它已略显过时。
“point the interested at the DLRM paper [1]”
“Check figure 6. of: Deep Learning Recommendation Model for Personalization and Recommendation Systems”
“An example is facebook DLRM: Deep Learning Recommendation Model for Personalization and Recommendation Systems”
“It's a little out of date, but the internal one is built with the same concepts, just more advanced modeling techniques and data.”