A Privacy-Preserved and Efficient Federated Learning Method Based on Important Data Selection.
Xiaoyin ZhaoPublished in: BMSB (2023)
Keyphrases
- synthetic data
- prior knowledge
- input data
- statistical methods
- unsupervised learning
- bayesian methods
- data analysis
- similarity measure
- information loss
- supervised learning
- test data
- knowledge discovery
- data sources
- data collection
- database
- computationally efficient
- clustering method
- data sets
- prior information
- noisy data
- sensitive information
- data mining techniques
- data points
- pairwise
- data structure
- objective function
- learning algorithm
- feature set
- active learning
- xml documents
- background knowledge
- learning process
- missing values
- preprocessing
- selection algorithm
- training data
- distributed data
- contingency tables