PPLL pipelines local learning modules across GPUs using queue-based communication, achieving 33-162% speedups over standard local learning and competitive accuracy on CIFAR-10, SVHN, and STL-10.
An analy- sis of single-layer networks in unsupervised feature learn- ing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Faster Multi-GPU Training with PPLL: A Pipeline Parallelism Framework Leveraging Local Learning
PPLL pipelines local learning modules across GPUs using queue-based communication, achieving 33-162% speedups over standard local learning and competitive accuracy on CIFAR-10, SVHN, and STL-10.