IIUM Repository

HELM based Reinforcement Learning for Goal Localization

AlDahoul, Nouar and Htike@Muhammad Yusof, Zaw Zaw (2016) HELM based Reinforcement Learning for Goal Localization. In: International Conference on Mechanical, Automotive and Aerospace Engineering, July 25-27, 2016, Kuala Lumpur, Malaysia. (Unpublished)

[img] PDF (Article)
Restricted to Repository staff only

Download (739kB) | Request a copy
[img] PDF (List of publications with paper IDs)
Restricted to Repository staff only

Download (44kB) | Request a copy
[img] PDF (Programme schedule)
Restricted to Repository staff only

Download (66kB) | Request a copy

Abstract

The objective of goal localizationHELM based reinforcement learning for goal localization is to find the location of goals in noisy environments. Simple actions are performed to move the agent towards the goal. The goal detector should be capable of minimizing the error between the predicted locations and the true ones. Few regions are processed by the agent to reduce the computational effort and increase the speed of convergence. In this paper, reinforcement learning method was utilized to find optimal series of actions to localize the goal region. The visual data, a set of images, is high dimensional unstructured data and needs to be represented efficiently to get a robust detector. Hierarchical Extreme Learning Machine (H-ELM) algorithm was used to find good features for effective representation. The results were analysed by using Matlab program

Item Type: Conference or Workshop Item (Other)
Additional Information: 6919/54994
Subjects: A General Works > AC Collections. Series. Collected works
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button): Kulliyyah of Engineering
Depositing User: Dr. Zaw Zaw Htike
Date Deposited: 08 Feb 2017 03:35
Last Modified: 08 Feb 2017 03:35
URI: http://irep.iium.edu.my/id/eprint/54994

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year