Predicting the Learning Rate of Gradient Descent for Accelerating Matrix Factorization.
Caio NóbregaLeandro Balby MarinhoPublished in: J. Inf. Data Manag. (2014)
Keyphrases
- learning rate
- matrix factorization
- error function
- stochastic gradient descent
- natural gradient
- collaborative filtering
- convergence rate
- recommender systems
- low rank
- learning algorithm
- negative matrix factorization
- weight vector
- missing data
- rapid convergence
- cost function
- adaptive learning rate
- multilayer neural networks
- nonnegative matrix factorization
- conjugate gradient
- factorization methods
- factor analysis
- latent factors
- item recommendation
- delta bar delta
- convergence speed
- convergence theorem
- objective function
- latent factor models
- evolutionary algorithm
- machine learning
- data representation