Solving MDPs with Skew Symmetric Bilinear Utility Functions
Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 1989-1994.
Abstract
In this paper we adopt Skew Symmetric Bilinear (SSB) utility functions to compare policies in Markov Decision Processes (MDPs). By considering pairs of alternatives, SSB utility theory generalizes von Neumann and Morgenstern's expected utility (EU) theory to encompass rational decision behaviors that EU cannot accommodate. We provide a game-theoretic analysis of the problem of identifying an SSB-optimal policy in finite horizon MDPs and propose an algorithm based on a double oracle approach for computing an optimal (possibly randomized) policy. Finally, we present and discuss experimental results where SSB-optimal policies are computed for a popular TV contest according to several instantiations of SSB utility functions.
@inproceedings{ijcai2015p282,
title = {Solving MDPs with Skew Symmetric Bilinear Utility Functions},
author = {Gilbert, Hugo and Spanjaard, Olivier and Viappiani, Paolo and Weng, Paul},
booktitle = {Proceedings of the Twenty-Fourth International Joint Conference on
Artificial Intelligence, {IJCAI-15}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Qiang Yang and Michael Wooldridge},
pages = {1989--1994},
year = {2015},
month = {7},
note = {Main Track},
url = {https://ijcai.org/proceedings/2015/282},
}
TY - CPAPER TI - Solving MDPs with Skew Symmetric Bilinear Utility Functions AU - Gilbert, Hugo AU - Spanjaard, Olivier AU - Viappiani, Paolo AU - Weng, Paul T2 - Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-15 PB - International Joint Conferences on Artificial Intelligence PY - 2015 SP - 1989 EP - 1994 UR - https://ijcai.org/proceedings/2015/282 ER -Copied
Machine-readable versions of this paper, for tools and AI agents: