Machine-imitative Learning by Using Computational Perceptions as Labeled Data-sets: A First Empirical Approximation

DOI

10.2991/asum.k.210827.053How to use a DOI?

Keywords

Computational Theory of Perceptions, Imitative Learning, Machine Learning, Intelligent Agents, Computer Games

Abstract

In this paper, we show how computational perceptions can be employed as labeled datasets to train agents in computer games. The idea is to automatically create a correspondence between perceptions and movements by using computational perception networks, next this knowledge is learned by the agents by using a decision tree. The result is a machine-imitative learning model able to mimic the human players. This approach is formally presented, a problem formulation based on the combination of linguistic descriptions of phenomena and classification is carried out. Additionally, we present a software architecture and module is explained. Finally, this architecture has been implemented and tested.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Clemente Rubio-Manzano
AU  - Tomás Lermanda
AU  - Claudia Martinez
AU  - Alejandra Segura
AU  - Christian Vidal
PY  - 2021
DA  - 2021/08/30
TI  - Machine-imitative Learning by Using Computational Perceptions as Labeled Data-sets: A First Empirical Approximation
BT  - Joint Proceedings of the 19th World Congress of the International Fuzzy Systems Association (IFSA), the 12th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT), and the 11th International Summer School on Aggregation Operators (AGOP)
PB  - Atlantis Press
SP  - 399
EP  - 404
SN  - 2589-6644
UR  - https://doi.org/10.2991/asum.k.210827.053
DO  - 10.2991/asum.k.210827.053
ID  - Rubio-Manzano2021
ER  -