2 citations · 4 across the 2 of their papers we have counts for
Showing cs.DCShow all
2 papers · 1 filter
cs.DC2021★ 2 cited
Adaptive Elastic Training for Sparse Deep Learning on Heterogeneous Multi-GPU Servers
Yujing Ma, Florin Rusu, Kesheng Wu +1
Motivated by extreme multi-label classification applications, we consider training deep learning models over sparse data in multi-GPU servers. The variance in the number of non-zer…
cs.DC2020★ 2 cited
Heterogeneous CPU+GPU Stochastic Gradient Descent Algorithms
Yujing Ma, Florin Rusu
The widely-adopted practice is to train deep learning models with specialized hardware accelerators, e.g., GPUs or TPUs, due to their superior performance on linear algebra operati…