Multistep Hyperparameter Tuning via Reinforcement Learning for Simulated Annealing.
Hiroshi YoshitakeTakuya OkuyamaYudai KamadaTaisuke UetaJunya FujitaPublished in: CoDIT (2023)
Keyphrases
- simulated annealing
- reinforcement learning
- simulated annealing algorithm
- genetic algorithm
- function approximation
- tabu search
- gaussian processes
- markov decision processes
- evolutionary algorithm
- metaheuristic
- solution quality
- model free
- hill climbing
- stochastic search
- learning algorithm
- reinforcement learning algorithms
- gaussian process
- genetic algorithm ga
- parameter settings
- optimization method
- machine learning
- rule selection
- temporal difference
- policy search
- bayesian networks
- reversible jump mcmc
- fine tuning
- global optimum
- hyperparameters
- optimal control
- kernel function
- feature selection