Výsledky vyhledávání - "Harm van Seijen"

On Value Function Representation of Long Horizon Problems

Autor: Lucas Lehnert, Romain Laroche, Harm Van Seijen

Publikováno v: Proceedings of the AAAI Conference on Artificial Intelligence. 32

In Reinforcement Learning, an intelligent agent has to make a sequence of decisions to accomplish a goal. If this sequence is long, then the agent has to plan over a long horizon. While learning the optimal policy and its value function is a well stu

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::f57c1d3de2ca540dafd76d9de287a610
https://doi.org/10.1609/aaai.v32i1.11646

Zobrazit plný text záznamu

EFFICIENT ABSTRACTION SELECTION IN REINFORCEMENT LEARNING

Autor: Shimon Whiteson, Harm van Seijen, Leon Kester

Publikováno v: Computational Intelligence. 30:657-699

This article addresses reinforcement learning problems based on factored Markov decision processes MDPs in which the agent must choose among a set of candidate abstractions, each build up from a different combination of state components. We present a

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::2d2b676af82ce76d4dbc7002b5850785
https://doi.org/10.1111/coin.12016

Zobrazit plný text záznamu

Exploiting best-match equations for efficient reinforcement learning

Autor: Harm van Seijen, Shimon Whiteson, Hado van Hasselt, Marco Wiering

Publikováno v: Journal of Machine Learning Research, 12, 2045-2094. Microtome Publishing
University of Groningen
Journal of Machine Learning Research, June, 12, 2045-2094
Journal of Machine Learning Research, 12(Jun), 2045-2094
Journal of Machine Learning Research, 12, 2045-2094

This article presents and evaluates best-match learning, a new approach to reinforcement learning that trades off the sample efficiency of model-based methods with the space efficiency of model-free methods. Best-match learning works by approximating

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::7f1843b3e6684827e53786b0a9b42ac3
https://dare.uva.nl/personal/pure/en/publications/exploiting-bestmatch-equations-for-efficient-reinforcement-learning(dfece862-a70f-4d0a-91eb-a0a9377ca1f4).html

Zobrazit plný text záznamu

HYBASE: hyperspectral band selection

Autor: Herman H. P. Th. Bekman, Harm van Seijen, Piet B. W. Schwering

Publikováno v: SPIE Proceedings.

Band selection is essential in the design of multispectral sensor systems. This paper describes the TNO hyperspectral band selection tool HYBASE. It calculates the optimum band positions given the number of bands and the width of the spectral bands.

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::019b544c459947bba32b4f5e3ada7883
https://doi.org/10.1117/12.818468

Zobrazit plný text záznamu

Postponed Updates for Temporal-Difference Reinforcement Learning

Autor: Harm van Seijen, Shimon Whiteson

Publikováno v: 9th International Conference on Intelligent Systems Design and Applications-ISDA'09, November 30-December 2, 2009, Pisa, Italy, 665-672
ISDA
2009 9th International Conference on Intelligent Systems Design and Applications (ISDA 2009): Pisa, Italy, 30 November-2 December 2009, 665-672
STARTPAGE=665;ENDPAGE=672;TITLE=2009 9th International Conference on Intelligent Systems Design and Applications (ISDA 2009): Pisa, Italy, 30 November-2 December 2009

This paper presents postponed updates, a new strategy for TD methods that can improve sample efficiency without incurring the computational and space requirements of model-based RL. By recording the agent's last-visit experience, the agent can delay

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1341e5d75e038dd48a92f65390bd938a
http://resolver.tudelft.nl/uuid:87daa5d4-0e9e-448e-af33-efa5d6c61997

Zobrazit plný text záznamu

A theoretical and empirical analysis of expected sarsa

Autor: Marco A. Wiering, Hado van Hasselt, Harm van Seijen, Shimon Whiteson

Publikováno v: 2009 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2009, 30 March-2 April 2009, Nashville, TN, USA, 177-184
Proceedings of the IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, 177-184
STARTPAGE=177;ENDPAGE=184;TITLE=Proceedings of the IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning
University of Groningen
Proceedings of the IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning: ADPRL
Proceedings of the IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning
ADPRL

This paper presents a theoretical and empirical analysis of Expected Sarsa, a variation on Sarsa, the classic onpolicy temporal-difference method for model-free reinforcement learning. Expected Sarsa exploits knowledge about stochasticity in the beha

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::13a5d0c2059407fd54b09a93fb7338a7
http://resolver.tudelft.nl/uuid:0e868206-5237-42d0-8bc0-3deb94a679ef

Zobrazit plný text záznamu

Texture analysis of coated conductors by Micro-Raman and synchrotron x-ray diffraction

Autor: A. Puig-Molina, Narcis Mestres, Alexander Usoskin, Herbert C. Freyhardt, Juan C. González, Harm van Seijen, Francesc Alsina, H. Graafsma, Teresa Puig, Xavier Obradors

Publikováno v: Scopus-Elsevier
Narcis Mestres Andreu

Two novel complementary and non-destructive techniques for texture analysis of YBCO coated conductors are presented. Micro-Raman (μ-Raman) spectroscopy enables an easy analysis of the film homogeneity by determining the distribution of a- and c-orie

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bf7af05d80672bf7e041a24371c4fb2b
http://www.scopus.com/inward/record.url?eid=2-s2.0-85010767265&partnerID=MN8TOARS

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání