AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Efficient, Secure, Differentially Private Deep Learning in the Two-Server Model

Jun Feng, Hong Sun, Pengfei Zhang, Bocheng Ren, Shunli Zhang

AAAI 2026 · Federated learning and privacy › Differential privacy · approach: Private training with public data · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace ReLU with bounded clipped activation (0 to 1)CDPDL-style DP training · MNIST (rho=8)test accuracy (%): 96.94 -> 97.25Improved
replace ReLU with bounded clipped activationCDPDL-style DP training · CIFAR10 (rho=17.5)test accuracy (%): 57.25 -> 58.63Improved
replace ReLU with bounded clipped activationCDPDL-style DP training · FashionMNIST (rho=2)test accuracy (%): 81.83 -> 83.02Improved
use garbled-circuits-free secure activation protocol instead of ABY2.0secure ReLU · runtime, batch 16total time (s): 1013.06 -> 88.6Improved
use garbled-circuits-free secure activation protocol instead of ABY2.0new activation · runtime, batch 16total time (s): 1859.15 -> 174.21Improved
use CryptDP secure gradient clipping and other protocols instead of ABY2.0-basedend-to-end training · runtimetotal time (s): 173960.9 -> 642.78Improved
use CryptDP instead of OUE/SUE local DP at much smaller privacy budgetCryptDP · FashionMNISTtest accuracy (%): 58.45 -> 86.11Improved
increase privacy loss rho from 2 to 10 (ablation)CryptDP · MNISTtest accuracy (%): 96.52 -> 97.42Improved
use CryptDP instead of OME-based LDPDLCryptDP · MNIST (rho=9)test accuracy (%): 95.26 -> 97.34Improved
use CryptDP softmax protocol instead of ABY2.0secure Softmax · runtime, batch 16total time (s): 357.32 -> 175.1Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy