Can we infer the presence of Differential Privacy in Deep Learning models' weights? Towards more secure Deep Learning.
Daniel Jiménez-LópezNuria Rodríguez BarrosoMaría Victoria LuzónFrancisco HerreraPublished in: CoRR (2023)
Keyphrases
- deep learning
- learning models
- differential privacy
- machine learning
- differentially private
- unsupervised feature learning
- privacy preserving
- unsupervised learning
- learning tasks
- loss function
- semi supervised learning
- learning algorithm
- machine learning algorithms
- data sharing
- privacy preservation
- learning problems
- classification models
- weakly supervised
- pattern recognition
- personal information
- decision trees
- sensitive data
- mental models
- computer vision
- data sets
- data mining
- training data
- conditional random fields
- text classification
- information extraction
- knowledge discovery
- active learning