Reinforcement Learning in Neural Networks: a Survey

Abstract

In recent years, researches on reinforcement learning (RL) have focused on bridging the gap between adaptive optimal control and bio-inspired learning techniques. Neural network reinforcement learning (NNRL) is among the most popular algorithms in the RL framework. The advantage of using neural networks enables the RL to search for optimal policies more efficiently in several real-life applications. Although many surveys investigated general RL, no survey is specifically dedicated to the combination of artificial neural networks and RL. This paper therefore describes the state of the art of NNRL algorithms, with a focus on robotics applications. In this paper, a comprehensive survey is started with a discussion on the concepts of RL. Then, a review of several different NNRL algorithms is presented. Afterwards, the performances of different NNRL algorithms are evaluated and compared in learning prediction and learning control tasks from an empirical aspect and the paper concludes with a discussion on open issues.

Authors and Affiliations

Ahmad Ghanbari| Faculty of Mechanical Engineering, and Mechatronics Research Laboratory, School of Engineering Emerging Technologies, University of Tabriz, Tabriz, Iran, Yasaman Vaghei| Mechatronics Research Laboratory, School of Engineering Emerging Technologies, University of Tabriz, Tabriz, Iran, Sayyed Mohammad Reza Sayyed Noorani| Mechatronics Research Laboratory, School of Engineering Emerging Technologies, University of Tabriz, Tabriz, Iran

Keywords

Related Articles

The First Report of Gastrointestinal Obstruction in New Zealand White Rabbit in Iran

Objective: Gastric dilatation also known as gastrointestinal obstruction or bloat is an acute and life-threatening condition in pet rabbits commonly caused by an intestinal obstruction with pellets, foreign bodies and...

Natural enemies of Ostrinia nubilalis (Hubner) and papulation density of pest and its dominant natural enemy on four corn hybrids in Moghan region

The European corn borer, Ostrinia nubilalis (Hübner), is an important insect pest of corn, Zea mays L., crop in Moghan region. In field experiments, the density of eggs and larvae of O.nubilalis and percentage of eggs...

Effect of processed lemon pulp with Saccharomyces cerevisiae yeast on protein and energy metabolism in Raini goats

The aim of present study is investigating effect of treated lemon pulp by Saccharomyces cerevisiae yeast on protein and energy metabolism in goats was fed with this product. In this experiment 8 goats from raini breed...

Cellulase Enzymes Production from Pistachio Hull by Different Strains of Tichoderma Fungi

Six strain of Trichoderma (T. viride, T. harzianum, T. reesei, T. longibrachiatum, T. koningii, T. virens) were isolated and used for extracellular enzyme production. Species of Trichoderma are collected from the surroun...

Farmer Behaviours and Sustainable Water Management in Semiarid Konya Closed Basin in Turkey

Objective: This study aims to review group learning method effect compared to individual learning method on dyslexic students of second grade in elementary school and it evaluates whether their problem will be solved i...

Download PDF file
  • EP ID EP13229
  • DOI -
  • Views 346
  • Downloads 12

How To Cite

Ahmad Ghanbari, Yasaman Vaghei, Sayyed Mohammad Reza Sayyed Noorani (2014). Reinforcement Learning in Neural Networks: a Survey. International journal of Advanced Biological and Biomedical Research, 2(5), 1398-1416. https://europub.co.uk/articles/-A-13229