Zobrazeno 1 - 10
of 22
pro vyhledávání: '"Lesner, Boris"'
Autor:
Lesner, Boris, Scherrer, Bruno
We consider approximate dynamic programming for the infinite-horizon stationary $\gamma$-discounted optimal control problem formalized by Markov Decision Processes. While in the exact case it is known that there always exists an optimal policy that i
Externí odkaz:
http://arxiv.org/abs/1304.5610
Autor:
Scherrer, Bruno, Lesner, Boris
Publikováno v:
NIPS 2012 (2012)
We consider infinite-horizon stationary $\gamma$-discounted Markov Decision Processes, for which it is known that there exists a stationary optimal policy. Using Value and Policy Iteration with some error $\epsilon$ at each iteration, it is well-know
Externí odkaz:
http://arxiv.org/abs/1211.6898
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Autor:
Lesner, Boris, Scherrer, Bruno
Publikováno v:
ICML 2015
ICML 2015, Jul 2015, Lille, France
ICML 2015, Jul 2015, Lille, France
International audience; We consider the infinite-horizon γ-discounted optimal control problem formalized by Markov Decision Processes. Running any instance of Modified Policy Iteration—a family of algorithms that can interpolate between Value and
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::8c57264d706d37319ca8eaae0abde0bf
https://inria.hal.science/hal-01186664
https://inria.hal.science/hal-01186664
Autor:
Crowell, Elizabeth Faris, Bazin, Cyril, Saunier, François, Brixtel, Romain, Caillot, Yann, Lesner, Boris, Toutain, Matthieu, Ferreri, Carole, Garcia, Incarnation, Mathieu, Marie-Claire, Vaussanvin, Jérôme, Depardon, Joëlle, Renouf, Arnaud
Publikováno v:
Acta Cytologica; 2019, Vol. 63 Issue 3, p215-223, 9p
Publikováno v:
Journal of Machine Learning Research
Journal of Machine Learning Research, 2015, 16, pp.1629−1676
Journal of Machine Learning Research, Microtome Publishing, 2015, 16, pp.1629−1676
Journal of Machine Learning Research, 2015, 16, pp.1629−1676
Journal of Machine Learning Research, Microtome Publishing, 2015, 16, pp.1629−1676
International audience; Modified policy iteration (MPI) is a dynamic programming (DP) algorithm that contains the two celebrated policy and value iteration methods. Despite its generality, MPI has not been thoroughly studied, especially its approxima
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::49672a67a0ab31164be4bbaeb3aa0902
https://inria.hal.science/hal-01091341/document
https://inria.hal.science/hal-01091341/document
Autor:
Scherrer, Bruno, Lesner, Boris
Publikováno v:
JFPDA-8èmes Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes-2013
JFPDA-8èmes Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes-2013, Jul 2013, Lille, France
JFPDA-8èmes Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes-2013, Jul 2013, Lille, France
National audience; Nous considérons les processus de décision Markoviens stationnaires et à horizon infini, pour lesquels on sait qu'il existe une politique optimale qui est stationnaire. En utilisant les algorithmes \emph{itérations sur les vale
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::d7d56dfbcfb4229e7c123980e50c7db8
https://hal.inria.fr/hal-00921291
https://hal.inria.fr/hal-00921291
Autor:
Lesner, Boris
Publikováno v:
Apprentissage [cs.LG]. université de caen, 2011. Français
We study Markovian Decision Processes represented with Probabilistic STRIPS action models.A first part of our works is about solving those processes in a compact way. To that end we proposetwo algorithms. A first one based on propositional formula ma
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::819050c6ac58d236a95b9ca93b43c082
https://hal.archives-ouvertes.fr/tel-01076437/file/These-Boris-Lesner-2011.pdf
https://hal.archives-ouvertes.fr/tel-01076437/file/These-Boris-Lesner-2011.pdf
Autor:
Lesner, Boris, Zanuttini, Bruno
Publikováno v:
Proc. 21st International Conference on Automated Planning and Scheduling (ICAPS 2011)
Proc. 21st International Conference on Automated Planning and Scheduling (ICAPS 2011))
Proc. 21st International Conference on Automated Planning and Scheduling (ICAPS 2011)), Jun 2011, Germany. 8 p
Proc. 21st International Conference on Automated Planning and Scheduling (ICAPS 2011))
Proc. 21st International Conference on Automated Planning and Scheduling (ICAPS 2011)), Jun 2011, Germany. 8 p
We present a novel dynamic programming approach to computing optimal policies for Markov Decision Processes compactly represented in grounded Probabilistic PDDL. Unlike other approaches, which use an intermediate representation as Dynamic Bayesian Ne
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f9b87994f8a41709a7c7dab5a7f7cc89
https://hal.archives-ouvertes.fr/hal-00944350/document
https://hal.archives-ouvertes.fr/hal-00944350/document