Learning a Machine for the Decision in a Partially Observable Markov Universe
Frederic Dambreville
Abstract
In this paper, we are interested in optimal decisions in a partially observable Markov universe. Our viewpoint departs from the dynamic programming viewpoint: we are directly approximating an optimal strategic tree depending on the observation. This approximation is made by means of a parameterized probabilistic law. In this paper, a particular family of hidden Markov models, with input and output, is considered as a learning framework. A method for optimizing the parameters of these HMMs is proposed and applied. This optimization method is based on the cross-entropic principle.
Create a lesson
Related papers
Even and Odd Integer Values of the Secondary Zeta Function
Artur Kawalec
Generalized Pellet's theorem for analytic matrix-valued functions
Hiromochi Itoh, Kei Matsushima, Takayuki Yamada
Certifying non-existence of integer points on curves
Bogdan Grechuk, Eugene Go
A finite coefficient theory for shape identification in tilted generalized gamma families
Ran Huang
There are only finitely many irregular Diophantine quadruples
Bo He, Alain Togbe
A Multipatch Model for Evaluating Interdivisional Connectivity and Vaccination Dynamics in the 2026 Bangladesh Measles Outbreak
Faizunnesa Khondaker, Md. Kamrujjaman