FedAT: A Communication-Efficient Federated Learning Method with Asynchronous Tiers under Non-IID Data.
Zheng ChaiYujing ChenLiang ZhaoYue ChengHuzefa RangwalaPublished in: CoRR (2020)
Keyphrases
- prior knowledge
- synthetic data
- input data
- statistical methods
- noisy data
- data sets
- test data
- computationally efficient
- data analysis
- bayesian methods
- unsupervised learning
- supervised learning
- data distribution
- missing data
- missing values
- data sources
- machine learning
- data collection
- neural network
- training data
- learning algorithm
- training samples
- prior information
- pairwise
- information loss
- learning process
- image data
- data acquisition
- database
- detection method
- background knowledge
- similarity measure
- clustering method
- semi supervised learning
- objective function
- feature set
- probabilistic model