Stochastic control with graphical models

the influence view approach

Paolo Magni, Riccardo Bellazzi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Markov decision processes (MDPs) allow to represent a wide class of problems in medical decision making and control. The complexity of the algorithms used to search the best policy of a MDP is directly related with the dimensionality of the state space. A careful structuring of the state space is hence an important task in the MDP specification. Graphical models are particularly appealing to cope with this task. In this paper we will describe a novel graphical formalism for MDP knowledge acquisition called Influence View (IV). An IV is a directed acyclic graph that depicts the probabilistic relationships between the problem state variables in a generic time transition; additional variables, called event variables, may be added, in order to describe the conditional independencies between state variables. By using the specified conditional independence structure, an IV may allow a parsimonious specification of a MDP. The authors have applied this methodology to the GVHD prophylaxis after Bone Marrow Transplantation.

Original languageEnglish
Title of host publicationAnnual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
PublisherIEEE
Pages1670-1671
Number of pages2
Volume4
Publication statusPublished - 1996
EventProceedings of the 1996 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Part 4 (of 5) - Amsterdam, Neth
Duration: Oct 31 1996Nov 3 1996

Other

OtherProceedings of the 1996 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Part 4 (of 5)
CityAmsterdam, Neth
Period10/31/9611/3/96

Fingerprint

Specifications
Knowledge acquisition
Bone
Decision making

ASJC Scopus subject areas

  • Bioengineering

Cite this

Magni, P., & Bellazzi, R. (1996). Stochastic control with graphical models: the influence view approach. In Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings (Vol. 4, pp. 1670-1671). IEEE.

Stochastic control with graphical models : the influence view approach. / Magni, Paolo; Bellazzi, Riccardo.

Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. Vol. 4 IEEE, 1996. p. 1670-1671.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Magni, P & Bellazzi, R 1996, Stochastic control with graphical models: the influence view approach. in Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. vol. 4, IEEE, pp. 1670-1671, Proceedings of the 1996 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Part 4 (of 5), Amsterdam, Neth, 10/31/96.
Magni P, Bellazzi R. Stochastic control with graphical models: the influence view approach. In Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. Vol. 4. IEEE. 1996. p. 1670-1671
Magni, Paolo ; Bellazzi, Riccardo. / Stochastic control with graphical models : the influence view approach. Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. Vol. 4 IEEE, 1996. pp. 1670-1671
@inproceedings{2d38af3eb5e34133b9efce86f1dc76ea,
title = "Stochastic control with graphical models: the influence view approach",
abstract = "Markov decision processes (MDPs) allow to represent a wide class of problems in medical decision making and control. The complexity of the algorithms used to search the best policy of a MDP is directly related with the dimensionality of the state space. A careful structuring of the state space is hence an important task in the MDP specification. Graphical models are particularly appealing to cope with this task. In this paper we will describe a novel graphical formalism for MDP knowledge acquisition called Influence View (IV). An IV is a directed acyclic graph that depicts the probabilistic relationships between the problem state variables in a generic time transition; additional variables, called event variables, may be added, in order to describe the conditional independencies between state variables. By using the specified conditional independence structure, an IV may allow a parsimonious specification of a MDP. The authors have applied this methodology to the GVHD prophylaxis after Bone Marrow Transplantation.",
author = "Paolo Magni and Riccardo Bellazzi",
year = "1996",
language = "English",
volume = "4",
pages = "1670--1671",
booktitle = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Stochastic control with graphical models

T2 - the influence view approach

AU - Magni, Paolo

AU - Bellazzi, Riccardo

PY - 1996

Y1 - 1996

N2 - Markov decision processes (MDPs) allow to represent a wide class of problems in medical decision making and control. The complexity of the algorithms used to search the best policy of a MDP is directly related with the dimensionality of the state space. A careful structuring of the state space is hence an important task in the MDP specification. Graphical models are particularly appealing to cope with this task. In this paper we will describe a novel graphical formalism for MDP knowledge acquisition called Influence View (IV). An IV is a directed acyclic graph that depicts the probabilistic relationships between the problem state variables in a generic time transition; additional variables, called event variables, may be added, in order to describe the conditional independencies between state variables. By using the specified conditional independence structure, an IV may allow a parsimonious specification of a MDP. The authors have applied this methodology to the GVHD prophylaxis after Bone Marrow Transplantation.

AB - Markov decision processes (MDPs) allow to represent a wide class of problems in medical decision making and control. The complexity of the algorithms used to search the best policy of a MDP is directly related with the dimensionality of the state space. A careful structuring of the state space is hence an important task in the MDP specification. Graphical models are particularly appealing to cope with this task. In this paper we will describe a novel graphical formalism for MDP knowledge acquisition called Influence View (IV). An IV is a directed acyclic graph that depicts the probabilistic relationships between the problem state variables in a generic time transition; additional variables, called event variables, may be added, in order to describe the conditional independencies between state variables. By using the specified conditional independence structure, an IV may allow a parsimonious specification of a MDP. The authors have applied this methodology to the GVHD prophylaxis after Bone Marrow Transplantation.

UR - http://www.scopus.com/inward/record.url?scp=0030312526&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030312526&partnerID=8YFLogxK

M3 - Conference contribution

VL - 4

SP - 1670

EP - 1671

BT - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

PB - IEEE

ER -