Empirical tests show differential privacy keeps latency and throughput close to normal while SMC depends on network speed and FHE adds roughly 1000x response time, with accuracy losses up to 35% on complex models.
Dif- ferential privacy techniques for cyber phys- ical systems: a survey [Journal Article]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Privacy-Preserving Machine Learning Framework for Edge Intelligence: An Empirical Analysis
Empirical tests show differential privacy keeps latency and throughput close to normal while SMC depends on network speed and FHE adds roughly 1000x response time, with accuracy losses up to 35% on complex models.