Skip to yearly menu bar Skip to main content


(4 events)   Timezone:  
Show all
Toggle Poster Visibility
Poster
Ballroom B - Position 43
RecD: Deduplication for End-to-End Deep Learning Recommendation Model Training Infrastructure
In Storage, Scheduling, and Networking
Mark Zhao · Dhruv Choudhary · Devashish Tyagi · Ajay Somani · Max Kaplan · Sung-Han Lin · Sarunya Pumma · Jongsoo Park · Aarti Basant · Niket Agarwal · Carole-Jean Wu · Christos Kozyrakis
[ Paper]  [ Slides]  [ Poster] 
Poster
Ballroom B - Position 41
μ-TWO: 3× Faster Multi-Model Training with Orchestration and Memory Optimization
In Storage, Scheduling, and Networking
Sanket Purandare · Abdul Wasay · Stratos Idreos · Animesh Jain
[ Paper]  [ Slides]  [ Poster] 
Poster
Ballroom B - Position 42
PyTorch RPC: Distributed Deep Learning Built on Tensor-Optimized Remote Procedure Calls
In Storage, Scheduling, and Networking
Pritam Damania · Shen Li · Alban Desmaison · Alisson Azzolini · Brian Vaughan · Edward Yang · Gregory Chanan · Guoqiang Jerry Chen · Hongyi Jia · Howard Huang · Joseph Spisak · Luca Wehrstedt · Lucas Hosseini · Manoj Krishnan · Omkar Salpekar · Pavel Belevich · Rohan Varma · Satendra Gera · Wanchao Liang · Shihao Xu · Soumith Chintala · Chaoyang He · Amir Ziashahabi · Salman Avestimehr · · Zachary DeVito
[ Paper]  [ Slides] 
Poster
Ballroom B - Position 40
Pre-train and Search: Efficient Embedding Table Sharding with Pre-trained Neural Cost Models
In Storage, Scheduling, and Networking
Daochen Zha · Louis Feng · Liang Luo · Bhargav Bhushanam · Zirui Liu · Yusuo Hu · Jade Nie · Yuzhen Huang · Yuandong Tian · Arun Kejariwal · Xia Hu
[ Paper]  [ Slides]  [ Poster]