activity
20162020
most citedDAPPLE: A Pipelined Data Parallel Approach for Training Large Models

29 citations · 39 across the 6 of their papers we have counts for

collaborators

8 papers

cs.DC20205 cited

Auto-MAP: A DQN Framework for Exploring Distributed Execution Plans for DNN Workloads

Siyu Wang, Yi Rong, Shiqing Fan +6

The last decade has witnessed growth in the computational requirements for training deep neural networks. Current approaches (e.g., data/model parallelism, pipeline parallelism) pa…

cs.DC202029 cited

DAPPLE: A Pipelined Data Parallel Approach for Training Large Models

Shiqing Fan, Yi Rong, Chen Meng +10

It is a challenging task to train large DNN models on sophisticated GPU platforms with diversified interconnect capabilities. Recently, pipelined training has been proposed as an e…

cs.DC2019

FusionStitching: Boosting Execution Efficiency of Memory Intensive Computations for DL Workloads

Guoping Long, Jun Yang, Wei Lin

Performance optimization is the art of continuous seeking a harmonious mapping between the application domain and hardware. Recent years have witnessed a surge of deep learning (DL…

cs.PF20195 cited

Characterizing Deep Learning Training Workloads on Alibaba-PAI

Mengdi Wang, Chen Meng, Guoping Long +4

Modern deep learning models have been exploited in various domains, including computer vision (CV), natural language processing (NLP), search and recommendation. In practical AI cl…

cs.LG2019

Efficient and Adaptive Kernelization for Nonlinear Max-margin Multi-view Learning

Changying Du, Jia He, Changde Du +3

Existing multi-view learning methods based on kernel function either require the user to select and tune a single predefined kernel or have to compute and store many Gram matrices…

cs.LG2019

Learning beyond Predefined Label Space via Bayesian Nonparametric Topic Modelling

Changying Du, Fuzhen Zhuang, Jia He +2

In real world machine learning applications, testing data may contain some meaningful new categories that have not been seen in labeled training data. To simultaneously recognize n…