Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling

Logic program and neural networks are two important aspects in artificial intelligence. This paper is part of an endeavour towards neural networks and logic programming integration. The goal in performing logic programming based on the energy minimization scheme is to achieve the best ratio of globa...

Full description

Bibliographic Details
Main Authors: Sathasivam, Saratha, Velavan, Muraly
Format: Conference or Workshop Item
Language:English
Published: 2018
Subjects:
Online Access:http://eprints.usm.my/41924/
http://eprints.usm.my/41924/1/Full_Paper_MAdurai.pdf
_version_ 1848879421269737472
author Sathasivam, Saratha
Velavan, Muraly
author_facet Sathasivam, Saratha
Velavan, Muraly
author_sort Sathasivam, Saratha
building USM Institutional Repository
collection Online Access
description Logic program and neural networks are two important aspects in artificial intelligence. This paper is part of an endeavour towards neural networks and logic programming integration. The goal in performing logic programming based on the energy minimization scheme is to achieve the best ratio of global minimum. However, there is no guarantee to find the best minimum in the network. To achieve this, activations functions are modified to accelerate the neuro symbolic integration. These activation functions will reduced the complexity of doing logic programming in Hopfield Neural Network (HNN).The activations functions discussed in this paper are new learning rule, Mc Culloch Pitts function and Hyperbolic Tangent Activation function. This paper also focused on agent based modelling for presenting performance of doing logic programming in Hopfield network using various activation functions. The effects of the activation function are analyzed mathematically and compared with the existing method. Computer simulations are carried out by using NETLOGO to validate the effectiveness on the new activation function. The resuls obtained showed that the Hyperbolic Tangent Activation function outperform other activation functions in doing logic programming in Hopfield network. The models developed by agent based modelling also support this theory.
first_indexed 2025-11-15T17:46:58Z
format Conference or Workshop Item
id usm-41924
institution Universiti Sains Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T17:46:58Z
publishDate 2018
recordtype eprints
repository_type Digital Repository
spelling usm-419242018-09-18T09:20:40Z http://eprints.usm.my/41924/ Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling Sathasivam, Saratha Velavan, Muraly QA1-939 Mathematics Logic program and neural networks are two important aspects in artificial intelligence. This paper is part of an endeavour towards neural networks and logic programming integration. The goal in performing logic programming based on the energy minimization scheme is to achieve the best ratio of global minimum. However, there is no guarantee to find the best minimum in the network. To achieve this, activations functions are modified to accelerate the neuro symbolic integration. These activation functions will reduced the complexity of doing logic programming in Hopfield Neural Network (HNN).The activations functions discussed in this paper are new learning rule, Mc Culloch Pitts function and Hyperbolic Tangent Activation function. This paper also focused on agent based modelling for presenting performance of doing logic programming in Hopfield network using various activation functions. The effects of the activation function are analyzed mathematically and compared with the existing method. Computer simulations are carried out by using NETLOGO to validate the effectiveness on the new activation function. The resuls obtained showed that the Hyperbolic Tangent Activation function outperform other activation functions in doing logic programming in Hopfield network. The models developed by agent based modelling also support this theory. 2018 Conference or Workshop Item PeerReviewed application/pdf en http://eprints.usm.my/41924/1/Full_Paper_MAdurai.pdf Sathasivam, Saratha and Velavan, Muraly (2018) Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling. In: International Conference on Applied and Pure Mathematics 2018, September 6-8, 2018, Madurai, TN, India. https://www.imrfedu.org/icm2018
spellingShingle QA1-939 Mathematics
Sathasivam, Saratha
Velavan, Muraly
Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling
title Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling
title_full Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling
title_fullStr Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling
title_full_unstemmed Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling
title_short Activation Functions in Neuro Symbolic Integration Using Agent Based Modelling
title_sort activation functions in neuro symbolic integration using agent based modelling
topic QA1-939 Mathematics
url http://eprints.usm.my/41924/
http://eprints.usm.my/41924/
http://eprints.usm.my/41924/1/Full_Paper_MAdurai.pdf