Brief

last 24h

[2/2] 223 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

TOOL · arXiv cs.LG English(EN) · 7h

DP-MacAdam: Differentially Private Mechanism with Adaptive Clipping and Adaptive Momentum

Researchers have introduced DP-MacAdam, a new algorithm designed to enhance privacy in machine learning training. This method combines adaptive clipping and adaptive momentum techniques, using the same gradient variance estimates for both processes. The algorithm aims to improve model utility over existing methods like DP-SGD and DP-Adam without requiring manual tuning of the clipping threshold. AI

IMPACT Introduces a novel algorithm for more effective privacy-preserving machine learning training.
RESEARCH · Hugging Face Daily Papers English(EN) · 2d · [3 sources]

Revisiting Privacy Amplification by Subsampling in Selective Release DPSGD

Researchers have identified a flaw in the privacy accounting of the Differentially Private Selective Update and Release (DPSUR) algorithm. The existing method overlooks variations in sampling probability introduced by its selective release mechanism, potentially weakening privacy guarantees. To address this, a new algorithm called Differentially Private Selective Release based on Clipped Gradients (DPSR-CG) has been proposed, which offers a more rigorous privacy analysis and demonstrates strong performance across various datasets. AI

IMPACT Enhances privacy guarantees for machine learning models trained on sensitive data, potentially enabling wider adoption in regulated industries.