REINFORCEMENT LEARNING IN COMPLEX REAL WORLD DOMAINS: A REVIEW
Journal Title: Indian Journal of Computer Science and Engineering - Year 2014, Vol 5, Issue 2
Abstract
Reinforcement Learning is an area of Machine Learning inspired by behaviorist psychology based on the mechanism of learning from rewards. RL does not require prior knowledge and automatically get optimal policy with the help of knowledge obtained by trial-and-error and continuous interaction with the dynamic environment. In complex real world domains implementing RL algorithms is the major practical problem due to the large and continuous space. It can give rise to problems like Curse of Dimensionality, Partial Observability Problem, Credit Structuring Problem, Generalization and Exploration-Exploitation Dilemma. This paper gives an introduction to Reinforcement Learning, discusses its basic model and system structure, and discusses the problems faced while implementing RL algorithms in complex real world domains. At last but not the least this paper briefly describes the techniques which can make the working of RL process easier in the complex domains. It concludes with research scope of RL in complex real world.
Authors and Affiliations
Samiksha Mahajan
Keywords
Related Articles
- EP ID EP162884
- DOI -
- Views 120
- Downloads 0
QOS PARAMETER ANALYSIS ON AODV AND DSDV PROTOCOLS IN A WIRELESS NETWORK
Wireless networks are characterized by a lack of infrastructure, and by a random and quickly changing network topology; thus the need for a robust dynamic routing protocol that can accommodate such an environment. To imp...
APPLICATION OF DATA MINING IN BIOINFORMATICS
This article highlights some of the basic concepts of bioinformatics and data mining. The major research areas of bioinformatics are highlighted. The application of data mining in the domain of bioinformatics is explaine...
Comparative analysis of Recurrent and Finite Impulse Response Neural Networks in Time Series Prediction
The purpose of this paper is to perform evaluation of two different neural network architectures used for solving temporal problems, i.e. time series prediction. The data sets in this project include Mackey-Glass, Sunspo...
A SURVEY ON FACE DETECTIONMETHODS AND FEATURE EXTRACTION OF FACE RECOGNITION USING PCA
From the most recent two decades, face acknowledgment is playing a vital and basic part particularly in the field of business, managing an account, social and law requirement region. It is an intriguing utilization of ex...
MINING OF ECG SIGNAL FOR NEW DIAGNOSTIC INFORMATION
This paper investigates a technique, which extracts new features having potentiality for giving more discriminatory clues from simple ECG (Electrocardiogram) signals. The extracted feature parameters, Average RR Signal M...