Introduces a repartitioning strategy that trades the number of data shuffles for variance reduction in large-scale distributed tuplewise estimation and learning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Trade-offs in Large-Scale Distributed Tuplewise Estimation and Learning
Introduces a repartitioning strategy that trades the number of data shuffles for variance reduction in large-scale distributed tuplewise estimation and learning.