Optimal Continuous Time Markov Decisions. Butkova, Y., Hatefi, H., Hermanns, H., & Krcal, J. arXiv:1507.02876 [cs], July, 2015. arXiv: 1507.02876Paper abstract bibtex In the context of Markov decision processes running in continuous time, one of the most intriguing challenges is the efficient approximation of finite horizon reachability objectives. A multitude of sophisticated model checking algorithms have been proposed for this. However, no proper benchmarking has been performed thus far. This paper presents a novel and yet simple solution: an algorithm originally developed for a restricted subclass of models and a subclass of schedulers can be twisted so as to become competitive with the more sophisticated algorithms in full generality. As the second main contribution, we perform a comparative evaluation of the core algorithmic concepts on an extensive set of benchmarks varying over all key parameters: model size, amount of non-determinism, time horizon, and precision.
@article{butkova_optimal_2015,
title = {Optimal {Continuous} {Time} {Markov} {Decisions}},
url = {http://arxiv.org/abs/1507.02876},
abstract = {In the context of Markov decision processes running in continuous time, one of the most intriguing challenges is the efficient approximation of finite horizon reachability objectives. A multitude of sophisticated model checking algorithms have been proposed for this. However, no proper benchmarking has been performed thus far. This paper presents a novel and yet simple solution: an algorithm originally developed for a restricted subclass of models and a subclass of schedulers can be twisted so as to become competitive with the more sophisticated algorithms in full generality. As the second main contribution, we perform a comparative evaluation of the core algorithmic concepts on an extensive set of benchmarks varying over all key parameters: model size, amount of non-determinism, time horizon, and precision.},
urldate = {2015-08-05},
journal = {arXiv:1507.02876 [cs]},
author = {Butkova, Yuliya and Hatefi, Hassan and Hermanns, Holger and Krcal, Jan},
month = jul,
year = {2015},
note = {arXiv: 1507.02876},
keywords = {Computer Science - Systems and Control},
}
Downloads: 0
{"_id":"v6cnZ8zTBA5nW4Zk8","bibbaseid":"butkova-hatefi-hermanns-krcal-optimalcontinuoustimemarkovdecisions-2015","downloads":0,"creationDate":"2015-07-07T15:41:10.117Z","title":"Optimal Continuous Time Markov Decisions","author_short":["Butkova, Y.","Hatefi, H.","Hermanns, H.","Krcal, J."],"year":2015,"bibtype":"article","biburl":"https://api.zotero.org/users/2539494/collections/QPKDCWZ2/items?key=GQZ7eSblSbBg8hlOjp5OpOAp&format=bibtex&limit=100","bibdata":{"bibtype":"article","type":"article","title":"Optimal Continuous Time Markov Decisions","url":"http://arxiv.org/abs/1507.02876","abstract":"In the context of Markov decision processes running in continuous time, one of the most intriguing challenges is the efficient approximation of finite horizon reachability objectives. A multitude of sophisticated model checking algorithms have been proposed for this. However, no proper benchmarking has been performed thus far. This paper presents a novel and yet simple solution: an algorithm originally developed for a restricted subclass of models and a subclass of schedulers can be twisted so as to become competitive with the more sophisticated algorithms in full generality. As the second main contribution, we perform a comparative evaluation of the core algorithmic concepts on an extensive set of benchmarks varying over all key parameters: model size, amount of non-determinism, time horizon, and precision.","urldate":"2015-08-05","journal":"arXiv:1507.02876 [cs]","author":[{"propositions":[],"lastnames":["Butkova"],"firstnames":["Yuliya"],"suffixes":[]},{"propositions":[],"lastnames":["Hatefi"],"firstnames":["Hassan"],"suffixes":[]},{"propositions":[],"lastnames":["Hermanns"],"firstnames":["Holger"],"suffixes":[]},{"propositions":[],"lastnames":["Krcal"],"firstnames":["Jan"],"suffixes":[]}],"month":"July","year":"2015","note":"arXiv: 1507.02876","keywords":"Computer Science - Systems and Control","bibtex":"@article{butkova_optimal_2015,\n\ttitle = {Optimal {Continuous} {Time} {Markov} {Decisions}},\n\turl = {http://arxiv.org/abs/1507.02876},\n\tabstract = {In the context of Markov decision processes running in continuous time, one of the most intriguing challenges is the efficient approximation of finite horizon reachability objectives. A multitude of sophisticated model checking algorithms have been proposed for this. However, no proper benchmarking has been performed thus far. This paper presents a novel and yet simple solution: an algorithm originally developed for a restricted subclass of models and a subclass of schedulers can be twisted so as to become competitive with the more sophisticated algorithms in full generality. As the second main contribution, we perform a comparative evaluation of the core algorithmic concepts on an extensive set of benchmarks varying over all key parameters: model size, amount of non-determinism, time horizon, and precision.},\n\turldate = {2015-08-05},\n\tjournal = {arXiv:1507.02876 [cs]},\n\tauthor = {Butkova, Yuliya and Hatefi, Hassan and Hermanns, Holger and Krcal, Jan},\n\tmonth = jul,\n\tyear = {2015},\n\tnote = {arXiv: 1507.02876},\n\tkeywords = {Computer Science - Systems and Control},\n}\n\n","author_short":["Butkova, Y.","Hatefi, H.","Hermanns, H.","Krcal, J."],"key":"butkova_optimal_2015","id":"butkova_optimal_2015","bibbaseid":"butkova-hatefi-hermanns-krcal-optimalcontinuoustimemarkovdecisions-2015","role":"author","urls":{"Paper":"http://arxiv.org/abs/1507.02876"},"keyword":["Computer Science - Systems and Control"],"metadata":{"authorlinks":{}}},"search_terms":["optimal","continuous","time","markov","decisions","butkova","hatefi","hermanns","krcal"],"keywords":["computer science - systems and control"],"authorIDs":["5457d87e2abc8e9f370007e5"],"dataSources":["ypi2bjGvbv5c7j2pz"]}