Parameter rollback averaged stochastic gradient descent for language model.
Zhao ChengGuanlin ChenWenyong WengQi LuWujian YangPublished in: J. Comput. Methods Sci. Eng. (2022)
Keyphrases
- language model
- stochastic gradient descent
- language modeling
- regularization parameter
- least squares
- loss function
- matrix factorization
- probabilistic model
- random forests
- step size
- mixture model
- information retrieval
- dirichlet prior
- weight vector
- online algorithms
- support vector machine
- logistic regression
- multiple kernel learning
- convergence rate
- image restoration
- cost function