all InfoSec news
Automatic Clipping: Differentially Private Deep Learning Made Easier and Stronger. (arXiv:2206.07136v1 [cs.LG])
June 16, 2022, 1:20 a.m. | Zhiqi Bu, Yu-Xiang Wang, Sheng Zha, George Karypis
cs.CR updates on arXiv.org arxiv.org
Per-example gradient clipping is a key algorithmic step that enables
practical differential private (DP) training for deep learning models. The
choice of clipping norm $R$, however, is shown to be vital for achieving high
accuracy under DP. We propose an easy-to-use replacement, called AutoClipping,
that eliminates the need to tune $R$ for any DP optimizers, including DP-SGD,
DP-Adam, DP-LAMB and many others. The automatic variants are as private and
computationally efficient as existing DP optimizers, but require no DP-specific
hyperparameters …
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
SOC 2 Manager, Audit and Certification
@ Deloitte | US and CA Multiple Locations
Regional Leader, Cyber Crisis Communications
@ Google | United Kingdom
Regional Intelligence Manager, Compliance, Safety and Risk Management
@ Google | London, UK
Senior Analyst, Endpoint Security
@ Scotiabank | Toronto, ON, CA, M1K5L1
Software Engineer, Security/Privacy, Google Cloud
@ Google | Bengaluru, Karnataka, India
Senior Security Engineer
@ Coinbase | Remote - USA