Privately Customizing Prefinetuning to Better Match User Data in Federated Learning.
Charlie HouHongyuan ZhanAkshat ShrivastavaSid WangSasha LivshitsGiulia FantiDaniel LazarPublished in: CoRR (2023)
Keyphrases
- prior knowledge
- data sources
- end users
- data sets
- database
- data processing
- data distribution
- data collection
- data analysis
- active learning
- spatial data
- learning systems
- data points
- probability distribution
- learning process
- data quality
- data structure
- reinforcement learning
- user defined
- high dimensional data
- synthetic data
- neural network
- user interests
- human experts
- log files
- distributed data
- background knowledge
- user interaction
- input data
- data mining techniques
- semi supervised
- high quality
- databases