Policy learning in continuous-time Markov decision processes using Gaussian Processes

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F17%3A00107689" target="_blank" >RIV/00216224:14330/17:00107689 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1016/j.peva.2017.08.007" target="_blank" >http://dx.doi.org/10.1016/j.peva.2017.08.007</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.peva.2017.08.007" target="_blank" >10.1016/j.peva.2017.08.007</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Policy learning in continuous-time Markov decision processes using Gaussian Processes

Popis výsledku v původním jazyce

Continuous-time Markov decision processes provide a very powerful mathematical framework to solve policy-making problems in a wide range of applications, ranging from the control of populations to cyber–physical systems. The key problem to solve for these models is to efficiently compute an optimal policy to control the system in order to maximise the probability of satisfying a set of temporal logic specifications. Here we introduce a novel method based on statistical model checking and an unbiased estimation of a functional gradient in the space of possible policies. Our approach presents several advantages over the classical methods based on discretisation techniques, as it does not assume the a-priori knowledge of a model that can be replaced by a black-box, and does not suffer from state-space explosion. The use of a stochastic moment-based gradient ascent algorithm to guide our search considerably improves the efficiency of learning policies and accelerates the convergence using the momentum term. We demonstrate the strong performance of our approach on two examples of non-linear population models: an epidemiology model with no permanent recovery and a queuing system with non-deterministic choice.

Název v anglickém jazyce

Policy learning in continuous-time Markov decision processes using Gaussian Processes

Popis výsledku anglicky

Continuous-time Markov decision processes provide a very powerful mathematical framework to solve policy-making problems in a wide range of applications, ranging from the control of populations to cyber–physical systems. The key problem to solve for these models is to efficiently compute an optimal policy to control the system in order to maximise the probability of satisfying a set of temporal logic specifications. Here we introduce a novel method based on statistical model checking and an unbiased estimation of a functional gradient in the space of possible policies. Our approach presents several advantages over the classical methods based on discretisation techniques, as it does not assume the a-priori knowledge of a model that can be replaced by a black-box, and does not suffer from state-space explosion. The use of a stochastic moment-based gradient ascent algorithm to guide our search considerably improves the efficiency of learning policies and accelerates the convergence using the momentum term. We demonstrate the strong performance of our approach on two examples of non-linear population models: an epidemiology model with no permanent recovery and a queuing system with non-deterministic choice.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA15-17564S" target="_blank" >GA15-17564S: Teorie her jako prostředek pro formální analýzu a verifikaci počítačových systémů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Performance Evaluation

ISSN

0166-5316

e-ISSN

—

Svazek periodika

116

Číslo periodika v rámci svazku

1

Stát vydavatele periodika

NL - Nizozemsko

Počet stran výsledku

17

Strana od-do

84-100

Kód UT WoS článku

000413797400005

EID výsledku v databázi Scopus

2-s2.0-85029590224