4 citations · 9 across the 5 of their papers we have counts for
Showing cs.DCShow all
2 papers · 1 filter
cs.DC2026
FCDP: Fully Cached Data Parallel for Communication-Avoiding Large-Scale Training
Gyeongseo Park, Eungyeong Lee, Song-woo Sok +5
Training billion-parameter models requires distributing model states across GPUs using fully sharded data parallel (i.e., ZeRO-3). While ZeRO-3 succeeds on clusters with high-bandw…
cs.DC2019★ 4 cited
NNStreamer: Stream Processing Paradigm for Neural Networks, Toward Efficient Development and Execution of On-Device AI Applications
MyungJoo Ham, Ji Joong Moon, Geunsik Lim +8
We propose nnstreamer, a software system that handles neural networks as filters of stream pipelines, applying the stream processing paradigm to neural network applications. A new…