Zobrazeno 1 - 10
of 14
pro vyhledávání: '"Yann-Michaël De Hauwere"'
Publikováno v:
The Knowledge Engineering Review. 31:59-76
Potential-based reward shaping is a commonly used approach in reinforcement learning to direct exploration based on prior knowledge. Both in single and multi-agent settings this technique speeds up learning without losing any theoretical convergence
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c3242dd085a7fdd9c45e49ed19b2dc8a
https://doi.org/10.1017/s0269888915000193
https://doi.org/10.1017/s0269888915000193
Autor:
Kristof Van Moffaert, Kevin Van Vaerenbergh, Yann-Michaël De Hauwere, Ann Nowé, Bruno Depraetere
Publikováno v:
SSCI
Heating a home is an energy consuming task. Most thermostats are programmed to turn on the heating at a particular time in order to reach and maintain a predefined target temperature. A lot of energy is wasted if these thermostats are not configured
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a479b8a7fde004fd1b26ac0cc864fc8b
https://doi.org/10.1109/ssci.2015.88
https://doi.org/10.1109/ssci.2015.88
Autor:
Thierry Salvant, Ivomar Brito Soares, Tim Brys, Kris Januarius, Yann-Michaël De Hauwere, Ann Nowé
Publikováno v:
ITSC
This paper considers how existing Reinforcement Learning (RL) techniques can be used to model and learn solutions for large scale Multi-Agent Systems (MAS). The large scale MAS of interest is the context of the movement of departure flights in big ai
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7fe9af35aa3358b6ac6ddf0f26f0b73f
https://doi.org/10.1109/itsc.2015.193
https://doi.org/10.1109/itsc.2015.193
Publikováno v:
Vrije Universiteit Brussel
When designing controllers for machines that interact with human users, it often becomes necessary to adapt control policies to user preferences, even when these preferences are not aligned with the optimal policy. In this paper, we present a reinfor
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::55240e7849e66abe4534a58044e440ca
https://biblio.vub.ac.be/vubir/adapting-control-policies-to-user-preferences(a5c58d6e-59ca-44ee-830e-bf29970298df).html
https://biblio.vub.ac.be/vubir/adapting-control-policies-to-user-preferences(a5c58d6e-59ca-44ee-830e-bf29970298df).html
Publikováno v:
Scopus-Elsevier
2012 Annual meeting of the North American Fuzzy Information Processing Society (NAFIPS 2012)
Vrije Universiteit Brussel
2012 Annual meeting of the North American Fuzzy Information Processing Society (NAFIPS 2012)
Vrije Universiteit Brussel
Satisfiability in propositional logic is well researched and many approaches to checking and solving exist. In infinite-valued or fuzzy logics, however, there have only recently been attempts at developing methods for solving satisfiability. In this
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::68f08e8d1321b507d2cad59b65d162c5
https://doi.org/10.1109/nafips.2012.6290998
https://doi.org/10.1109/nafips.2012.6290998
Publikováno v:
Vrije Universiteit Brussel
Adaptive and Learning Agents ISBN: 9783642284984
ALA
Adaptive and Learning Agents ISBN: 9783642284984
ALA
One of the main advantages of Reinforcement Learning is the capability of dealing with a delayed reward signal. Using an appropriate backup diagram, rewards are backpropagated through the state space. This allows agents to learn to take the correct a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d240aed5f4f96bc7d4c395b849c1ac72
https://biblio.vub.ac.be/vubir/solving-sparse-delayed-coordination-problems-in-multiagent-reinforcement-learning(6ebe99db-730b-4e6e-bf96-b30bd82b01e3).html
https://biblio.vub.ac.be/vubir/solving-sparse-delayed-coordination-problems-in-multiagent-reinforcement-learning(6ebe99db-730b-4e6e-bf96-b30bd82b01e3).html
Publikováno v:
Adaptation, Learning, and Optimization ISBN: 9783642276446
Reinforcement Learning
Vrije Universiteit Brussel
Reinforcement Learning
Vrije Universiteit Brussel
Reinforcement Learning was originally developed for Markov Decision Processes (MDPs). It allows a single agent to learn a policy that maximizes a possibly delayed reward signal in a stochastic stationary environment. It guarantees convergence to the
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3576fd397b8a21387924c9a864f311c4
https://biblio.vub.ac.be/vubir/game-theory-and-multiagent-reinforcement-learning(528d8fe4-c46e-44dd-958b-49b1f8f71259).html
https://biblio.vub.ac.be/vubir/game-theory-and-multiagent-reinforcement-learning(528d8fe4-c46e-44dd-958b-49b1f8f71259).html
Publikováno v:
Vrije Universiteit Brussel
Recent research has demonstrated that considering local interactions among agents in specific parts of the state space, is a successful way of simplifying the multi-agent learning process. By taking into account other agents only when a conflict is p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::10dc0382b1c578e2781b21c37d34feb4
https://hdl.handle.net/20.500.14017/fb9da028-9e5e-471e-8fe6-4fdcaa7fce96
https://hdl.handle.net/20.500.14017/fb9da028-9e5e-471e-8fe6-4fdcaa7fce96
A major challenge in multi-agent reinforcement learning remains dealing with the large state spaces typically associated with realistic multi-agent systems. As the state space grows, agent policies become increasingly complex and learning slows down.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::50d80c89142bd1632fff64306763fd04
https://hdl.handle.net/20.500.14017/418f0c69-8618-4ea3-9d44-9e8526f3e527
https://hdl.handle.net/20.500.14017/418f0c69-8618-4ea3-9d44-9e8526f3e527
Publikováno v:
Agent and Multi-agent Technology for Internet and Enterprise Systems ISBN: 9783642135255
Agent and Multi-agent Technology for Internet and Enterprise Systems
Vrije Universiteit Brussel
Agent and Multi-agent Technology for Internet and Enterprise Systems
Vrije Universiteit Brussel
A major challenge in multi-agent reinforcement learning remains dealing with the large state spaces typically associated with realistic multi-agent systems. As the state space grows, agent policies become more and more complex and learning slows down
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9e68df899ebf83be7e881c8cf1f1ccab
https://doi.org/10.1007/978-3-642-13526-2_9
https://doi.org/10.1007/978-3-642-13526-2_9