Adaptive Proposer for Ultimatum Game

0462888 - ÚTIA 2017 RIV CH eng C - Conference Paper (international conference)
Hůla, František - Ruman, Marko - Kárný, Miroslav
Adaptive Proposer for Ultimatum Game.
Artificial Neural Networks and Machine Learning – ICANN 2016. Vol. Part I. Cham: Springer, 2016, s. 330-338. Lecture Notes in Computer Science, 9886. ISBN 978-3-319-44777-3. ISSN 0302-9743.
[International Conference on Artificial Neural Networks 2016 /25./. Barcelona (ES), 06.09.2016-09.09.2016]
R&D Projects: GA ČR GA13-13502S
Institutional support: RVO:67985556
Keywords : Games * Markov decision process * Bayesian learning
Subject RIV: BB - Applied Statistics, Operational Research
http://library.utia.cas.cz/separaty/2016/AS/karny-0462888.pdf

Ultimate Game serves for extensive studies of various aspects of human decision making. The current paper contribute to them by designing proposer optimising its policy using Markov-decision-process (MDP) framework combined with recursive Bayesian learning of responder’s model. Its foreseen use: i) standardises experimental conditions for studying rationality and emotion-influenced decision making of human responders; ii) replaces the classical game-theoretical design of the players’ policies by an adaptive MDP, which is more realistic with respect to the knowledge available to individual players and decreases player’s deliberation effort; iii) reveals the need for approximate learning and dynamic programming inevitable for coping with the curse of dimensionality; iv) demonstrates the influence of the fairness attitude of the proposer on the game course; v) prepares the test case for inspecting exploration-exploitation dichotomy.
Permanent Link: http://hdl.handle.net/11104/0262368