Markov Indecision Processes: A Formal Model of Decision-Making under Extreme Confusion

Abstract:

»We present a mathematical model of indecisive agents faced with a sequence of diffcult decisions, extending Adams‘ bistromathics to the multistage case. This is almost the first work on modeling stochastic processes for which the probabilities are fundamentally unknowable. This paper describes a novel algorithm, complexity results, and a model-free learning algorithm for Markov indecision processes. Two applications are discussed based on real-world domains: presidential elections and the stock market.«