Negotiations in strategic economic games – challenges for developing an artificial intelligence engine
Justyna Majewska , Szymon Truskolaski
AbstractThe aim of the paper is twofold: 1) to assess the existing engine mechanisms and quality of negotiations in strategic economic games within Player‐NPC interactions by determining the gap in delivering realistic gameplay, and 2) to propose a design of an multi‐ agent system based on artificial intelligence providing a new quality of negotiations in strategy games – more realistic, ie. replicating behaviour of “live” players, enhancing the useability of the game (satisfaction of the human player in the single player game). For this purpose, an in‐ depth analysis of the literature on the subject and evaluation of the applicability of multi‐ agent systems and AI techniques, such as artificial neural networks, Monte Carlo tree search (MCTS), reinforcement learning (Q‐learning) or fuzzy logic will be carried out in order to develop a framework that suport machine learning model. As a result we obtain a model of interactions between agents (modules) in strategic economic games, being intelligent multi‐ agent system, that are more realistic by replicating behaviour of human players within Player‐ NPC interactions and reproducing the complexity of the relationships that occur in socioeconomic systems – according to the paradigms of modern economic systems that increases the complexity of economic processes.
|Publication size in sheets||0.6|
|Book||Tomé Eduardo, Neumann Gaby, Majewska Justyna, Truskolaski Szymon (eds.): Proceedings of the International Conference Theory and Applications in the Knowledge Economy TAKE 2018 – Poznan, Poland, 11 to 13 July 2018, 2018, Uniwersytet Ekonomiczny w Poznaniu, ISBN 978-999-54182-0-6, 532 p.|
|Keywords in Polish||realistyczna rozgrywka, negocjacje, ekonomiczne gry strategiczne, silnik AI, system wieloagentowy|
|Keywords in English||realistic gameplay, negotiations, economic strategy games, AI engine, multi‐agents system|
|Score||= 20.0, 23-04-2020, ChapterFromConference|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.