Lottery Hypothesis based Unsupervised Pre-training for Model Compression in Federated Learning.
Sohei ItaharaTakayuki NishioMasahiro MorikuraKoji YamamotoPublished in: CoRR (2020)
Keyphrases
- learning scheme
- learning algorithm
- prior knowledge
- computational model
- supervised learning
- learning process
- active learning
- probabilistic model
- mathematical model
- high level
- structured prediction
- learning mechanism
- weakly supervised
- training process
- graphical models
- hidden markov models
- learning systems
- unsupervised learning
- probability distribution
- data sources
- digital libraries
- reinforcement learning
- learning speed