Momentum-accelerated Frank-Wolfe and gradient descent reduce both iteration counts and privacy noise for private and heavy-tailed-robust estimation, yielding rates such as 1/(nε) instead of 1/(nε)^{2/3}.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.ST 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Benefits of Accelerated Optimization in Robust and Private Estimation
Momentum-accelerated Frank-Wolfe and gradient descent reduce both iteration counts and privacy noise for private and heavy-tailed-robust estimation, yielding rates such as 1/(nε) instead of 1/(nε)^{2/3}.