Parallelization of the self-organized maps algorithm for federated learning on distributed sources.
Ivan KholodAndrey RukavitsynAlexey A. PaznikovSergei GorlatchPublished in: J. Supercomput. (2021)
Keyphrases
- learning algorithm
- experimental evaluation
- k means
- dynamic programming
- times faster
- clustering method
- evolutionary algorithm
- high accuracy
- learning speed
- detection algorithm
- theoretical analysis
- optimization algorithm
- similarity measure
- distributed data
- reinforcement learning
- path planning
- preprocessing
- cost function
- learning process
- online learning
- expectation maximization
- input data
- simulated annealing
- objective function
- prior knowledge
- worst case
- learning tasks
- supervised learning
- learning phase
- distributed learning
- data sources