Support of imperfect decision-making agents
| advisor: | Ing. Miroslav Kárný, DrSc. |
| e-mail: | show e-mail |
| type: | phd thesis |
| branch of study: | MI_MM, MI_AMSM |
| key words: | dynamic decision making; uncertainty; decision support |
| description: | Real (imperfect) decision-making (DM) agents can employ advanced, near-optimal, decision policies only if they can prepare the quantitative inputs into the employed policy design. Essentially, they specify: a) belief model of the environment that may contain other DM agents; b) their DM preferences. The specification is a hard problem on its own. It requires to solve a range of specific sub- and meta- DM-tasks. Thesis will focus on a selected important task of this type. Its proper specification and solution will form the thesis core. The prescriptive DM theory known as fully probabilistic design [1,2], which extends Bayesian DM [3,4], combined [5] with a rich knowledge accumulated by descriptive DM theories [6,7] will form the targeted DM to be supported. |
| references: | The literature will be gradually enriched according to the progress and needs.
[1] M. Kárný: Axiomatisation of Fully Probabilistic Design Revisited, Systems and Control Letters 141, (2020) [2] M. Kárný, J. Böhm, T.V. Guy et al: Optimized Bayesian Dynamic Advising: Theory and Algorithms, Springer, London (2006) [3] M.H. DeGroot: Optimal Statistical Decisions, McGraw-Hill, (1970) [4] J.C. Harsanyi: Games with Incomplete Information Played by Bayesian Players I-III, Management Science, 50(12), Suppl, (2004) [5] T.V. Guy, M. Kárný, D. H. Wolpert: Decision Making: Uncertainty, Imperfection, Deliberation and Scalability, Springer, Cham, Studies in Computational Intelligence 538, (2014) [6] D. Kahneman: Thinking, fast and slow, Penguin, (2011) [7] G. Fowler: How emotional AI is creating personalized customer experiences and making a social impact, Forbes, (2019) last update: 6/12/21 |
| last update: | 06.12.2021 15:31:52 |
administrator for this page:
Ľubomíra Dvořáková | last update: 09/12/2011