Home // COGNITIVE 2019, The Eleventh International Conference on Advanced Cognitive Technologies and Applications // View article

Neural Network-Based Reasoning for Solving the Tower of Hanoi

Authors:
Sanghun Bang
Charles Tijus

Keywords: Reasoning;Inference;Neural Networks; Machine Learning;Tower of Hanoi.

Abstract:
In this paper, we propose a reasoning solution, which can infer the rules of game, such as Tower of Hanoi. Neural networks require large amounts of data to improve performance. However, human intelligence requires only a simple exposition. The goal of this paper is to learn to solve a Tower of Hanoi without much learning. We collect sequential data from participants’ experiments that are supposed to solve the Tower of Hanoi. And then, we observe relations between different objects and their actions to establish our reasoning model. We train our reasoning model to solve the Tower of Hanoi. Finally, we show that our reasoning model can infer the rule of game through observations from objects, their relationships and actions.

Pages: 83 to 84

Copyright: Copyright (c) IARIA, 2019

Publication date: May 5, 2019

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-705-4

Location: Venice, Italy

Dates: from May 5, 2019 to May 9, 2019