On the convergence of the MLE as an estimator of the learning rate in the Exp3 algorithm.
Julien AubertLuc LehéricyPatricia Reynaud-BouretPublished in: CoRR (2023)
Keyphrases
- convergence rate
- rapid convergence
- learning rate
- learning algorithm
- convergence theorem
- high accuracy
- iterative algorithms
- dynamic programming
- adaptive learning rate
- maximum likelihood
- cost function
- faster convergence
- optimization algorithm
- optimal solution
- error function
- weight update
- levenberg marquardt
- data mining
- linear programming
- least squares
- objective function
- training algorithm
- em algorithm
- expectation maximization
- maximum likelihood estimator
- update rule
- machine learning
- neural network
- delta bar delta