Probing the Transition to Dataset-Level Privacy in ML Models Using an Output-Specific and Data-Resolved Privacy Profile.
Tyler LeBlondJoseph MunozFred LuMaya FuchsElliott Zaresky-WilliamsEdward RaffBrian TestaPublished in: CoRR (2023)
Keyphrases
- experimental data
- private data
- database
- sensitive information
- personal data
- data privacy
- data sets
- original data
- data collection
- data processing
- input data
- image data
- knowledge discovery
- private information
- data structure
- information loss
- privacy issues
- statistical methods
- historical data
- privacy preservation
- statistical databases
- differentially private
- data quality
- spatial data
- synthetic data
- personal information
- model selection
- prior knowledge
- decision trees
- training data
- preserving privacy
- individual privacy
- privacy requirements
- feature selection
- data protection
- social network data
- differential privacy
- learned models
- data analysis
- data sources
- sensitive data
- training dataset
- higher level