Pith. sign in

Private empirical risk minimization: Efficient algorithms and tight error bounds

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

On Design Principles for Private Adaptive Optimizers

cs.LG · 2025-07-01 · conditional · novelty 5.0

A theoretical and empirical study finds that unbiased second-moment estimates in private Adam can be harmful in high dimensions, and that scale-then-privatize outperforms the alternatives on a small transformer task.

citing papers explorer

Showing 1 of 1 citing paper.

  • On Design Principles for Private Adaptive Optimizers cs.LG · 2025-07-01 · conditional · none · ref 4

    A theoretical and empirical study finds that unbiased second-moment estimates in private Adam can be harmful in high dimensions, and that scale-then-privatize outperforms the alternatives on a small transformer task.