SK converges linearly at rate 1-σ2 once near the optimum, with explicit burn-in, and can be accelerated by gradient methods on the semi-dual.
Estimating nonnegative matrices from marginal data.International Economic Review, 6(3):294–310, 1965
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Tight Nonasymptotic Local Convergence of Sinkhorn-Knopp
SK converges linearly at rate 1-σ2 once near the optimum, with explicit burn-in, and can be accelerated by gradient methods on the semi-dual.