Curse or Redemption? How Data Heterogeneity Affects the Robustness of Federated Learning.
Syed ZawadAhsan AliPin-Yu ChenAli AnwarYi ZhouNathalie BaracaldoYuan TianFeng YanPublished in: CoRR (2021)
Keyphrases
- data points
- data sets
- synthetic data
- learning algorithm
- hidden variables
- raw data
- high quality
- data sources
- high dimensional data
- distributed data
- database
- learning tasks
- experimental data
- background knowledge
- knowledge acquisition
- decision trees
- training data
- data analysis
- small number
- knowledge discovery
- neural network
- bayesian methods
- original data
- data structure
- digital libraries
- prior knowledge
- training examples
- information sources
- computer systems
- data collection
- image data
- data processing
- online learning
- input data