English

Towards Autonomous Reinforcement Learning: Automatic Setting of Hyper-parameters using Bayesian Optimization

Artificial Intelligence 2018-05-15 v1 Machine Learning

Abstract

With the increase of machine learning usage by industries and scientific communities in a variety of tasks such as text mining, image recognition and self-driving cars, automatic setting of hyper-parameter in learning algorithms is a key factor for achieving satisfactory performance regardless of user expertise in the inner workings of the techniques and methodologies. In particular, for a reinforcement learning algorithm, the efficiency of an agent learning a control policy in an uncertain environment is heavily dependent on the hyper-parameters used to balance exploration with exploitation. In this work, an autonomous learning framework that integrates Bayesian optimization with Gaussian process regression to optimize the hyper-parameters of a reinforcement learning algorithm, is proposed. Also, a bandits-based approach to achieve a balance between computational costs and decreasing uncertainty about the Q-values, is presented. A gridworld example is used to highlight how hyper-parameter configurations of a learning algorithm (SARSA) are iteratively improved based on two performance functions.

Keywords

Cite

@article{arxiv.1805.04748,
  title  = {Towards Autonomous Reinforcement Learning: Automatic Setting of Hyper-parameters using Bayesian Optimization},
  author = {Juan Cruz Barsce and Jorge A. Palombarini and Ernesto C. Martínez},
  journal= {arXiv preprint arXiv:1805.04748},
  year   = {2018}
}

Comments

Paper submitted to CLEI Electronic Journal. This is an extended version of the conference paper presented at Latin American Computer Conference (CLEI), 2017

R2 v1 2026-06-23T01:52:56.742Z