Fed-CBS: A Heterogeneity-Aware Client Sampling Mechanism for Federated Learning via Class-Imbalance Reduction.
Jianyi ZhangAng LiMinxue TangJingwei SunXiang ChenFan ZhangChangyou ChenYiran ChenHai LiPublished in: ICML (2023)
Keyphrases
- class imbalance
- learning process
- active learning
- learning algorithm
- pattern recognition
- prior knowledge
- imbalanced data sets
- sampling methods
- learning tasks
- unsupervised learning
- imbalanced class distribution
- classification error
- high dimensionality
- text categorization
- semi supervised
- support vector machine
- reinforcement learning