On the convergence of the MLE as an estimator of the learning rate in the Exp3 algorithm.
Julien AubertLuc LehéricyPatricia Reynaud-BouretPublished in: ICML (2023)
Keyphrases
- learning rate
- convergence rate
- rapid convergence
- convergence theorem
- learning algorithm
- adaptive learning rate
- high accuracy
- dynamic programming
- maximum likelihood
- particle swarm optimization
- convergence speed
- error function
- data points
- least squares
- expectation maximization
- optimization algorithm
- pairwise
- iterative algorithms
- optimal solution
- natural gradient
- reinforcement learning
- neural network