Enhancing Communication Efficiency and Training Time Uniformity in Federated Learning through Multi-Branch Networks and the Oort Algorithm.
Pin-Hung JuanJa-Ling WuPublished in: Algorithms (2024)
Keyphrases
- learning algorithm
- learning speed
- computational complexity
- objective function
- high efficiency
- incremental learning
- computational efficiency
- learning phase
- optimal solution
- training phase
- np hard
- particle swarm optimization
- detection algorithm
- preprocessing
- matching algorithm
- highly efficient
- structured prediction
- feedforward neural networks
- fully connected
- stochastic gradient descent
- contrastive divergence
- learning stage
- recurrent networks
- neural network
- training process
- supervised learning
- data sources
- computational cost
- mobile robot
- cost function
- evolutionary algorithm
- artificial neural networks
- image segmentation
- decision trees
- social networks
- genetic algorithm