Privacy Amplification via Compression: Achieving the Optimal Privacy-Accuracy-Communication Trade-off in Distributed Mean Estimation.
Wei-Ning ChenDan SongAyfer ÖzgürPeter KairouzPublished in: NeurIPS (2023)
Keyphrases
- trade off
- privacy preserving
- communication overhead
- personal information
- computational cost
- privacy preserving data mining
- estimation accuracy
- privacy concerns
- error analysis
- data privacy
- multi party
- information loss
- estimation error
- communication cost
- private data
- dynamic programming
- preserving privacy
- privacy preservation
- statistical databases
- private information
- spatially distributed
- security issues
- distributed systems
- distributed control
- personal data
- image compression
- location privacy
- exhaustive search
- computer networks
- privacy enhancing
- security risks
- sensor networks
- fully distributed
- distributed computation
- privacy issues
- differential privacy
- compression scheme
- error rate
- prediction accuracy
- distributed data
- computational complexity
- distributed environment
- data compression