A fixed-support SVD tensor factorization with tiling and bipartite matching yields an explicit zero-error achievable rate K/N for multi-user non-linear distributed computing under mild dimensionality conditions.
Heterogeneity matters even more in distributed learning: Study from generalization perspective,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IT 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Multi-User Non-Linearly Separable Distributed Computing
A fixed-support SVD tensor factorization with tiling and bipartite matching yields an explicit zero-error achievable rate K/N for multi-user non-linear distributed computing under mild dimensionality conditions.