213 citations · 240 across the 12 of their papers we have counts for
3 papers · 1 filter
Mimose: An Input-Aware Checkpointing Planner for Efficient Training on GPU
Jianjin Liao, Mingzhen Li, Qingxiao Sun +8
Larger deep learning models usually lead to higher model quality with an ever-increasing GPU memory footprint. Although tensor checkpointing techniques have been proposed to enable…
EasyScale: Accuracy-consistent Elastic Training for Deep Learning
Mingzhen Li, Wencong Xiao, Biao Sun +9
Distributed synchronized GPU training is commonly used for deep learning. The resource constraint of using a fixed number of GPUs makes large-scale training jobs suffer from long q…
FamilySeer: Towards Optimized Tensor Codes by Exploiting Computation Subgraph Similarity
Shanjun Zhang, Mingzhen Li, Hailong Yang +3
Deploying various deep learning (DL) models efficiently has boosted the research on DL compilers. The difficulty of generating optimized tensor codes drives DL compiler to ask for…