Zobrazeno 1 - 10
of 84
pro vyhledávání: '"Haitham Bou-Ammar"'
Autor:
Alexander I. Cowen-Rivers, Wenlong Lyu, Rasul Tutunov, Zhi Wang, Antoine Grosnit, Ryan Rhys Griffiths, Alexandre Max Maraval, Hao Jianye, Jun Wang, Jan Peters, Haitham Bou-Ammar
Publikováno v:
Journal of Artificial Intelligence Research. 74:1269-1349
In this work we rigorously analyse assumptions inherent to black-box optimisation hyper-parameter tuning tasks. Our results on the Bayesmark benchmark indicate that heteroscedasticity and non-stationarity pose significant challenges for black-box opt
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a97fe53f0c70f1b70dc6e8de08640b0e
https://doi.org/10.1613/jair.1.13643
https://doi.org/10.1613/jair.1.13643
Autor:
Alexander I. Cowen-Rivers, Daniel Palenicek, Vincent Moens, Mohammed Amin Abdullah, Aivar Sootla, Jun Wang, Haitham Bou-Ammar
Publikováno v:
Machine Learning. 111:173-203
In this paper, we propose SAMBA, a novel framework for safe reinforcement learning that combines aspects from probabilistic modelling, information theory, and statistics. Our method builds upon PILCO to enable active exploration using novel(semi-)met
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6c698c64f06f947b174b6bf355fc68d3
https://doi.org/10.1007/s10994-021-06103-6
https://doi.org/10.1007/s10994-021-06103-6
Autor:
Asif Khan, Alexander Imani Cowen-Rivers, Derrick-Goh-Xin Deik, Antoine Grosnit, Philippe ROBERT, Victor Greiff, Eva Smorodina, Puneet Rawat, Rahmad Akbar, Kamil Dreczkowski, Rasul Tatunov, Dany Bou-Ammar, Jun Wang, Haitham Bou-Ammar
Publikováno v:
SSRN Electronic Journal.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a514d4ffc22739654ff39aa27ae7b79e
https://doi.org/10.2139/ssrn.4115860
https://doi.org/10.2139/ssrn.4115860
Autor:
Asif Khan, Alexander I. Cowen-Rivers, Antoine Grosnit, Derrick-Goh-Xin Deik, Philippe A. Robert, Victor Greiff, Eva Smorodina, Puneet Rawat, Rahmad Akbar, Kamil Dreczkowski, Rasul Tutunov, Dany Bou-Ammar, Jun Wang, Amos Storkey, Haitham Bou-Ammar
Publikováno v:
Cell Reports Methods. 3:100374
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3b05b66c34a9da60efb06d45ef8b7f11
https://doi.org/10.1016/j.crmeth.2022.100374
https://doi.org/10.1016/j.crmeth.2022.100374
Publikováno v:
IEEE Transactions on Automatic Control. 64:3983-3994
In this paper, we propose a distributed Newton method for decenteralized optimization of large sums of convex functions. Our proposed method is based on creating a set of separable finite sum minimization problems by utilizing a decomposition techniq
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::82705a1a48ae78dae35237a243b49275
https://doi.org/10.1109/tac.2019.2907711
https://doi.org/10.1109/tac.2019.2907711
Highly dynamic robotic tasks require high-speed and reactive robots. These tasks are particularly challenging due to the physical constraints, hardware limitations, and the high uncertainty of dynamics and sensor measures. To face these issues, it's
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7992ebadf07c335c828ea7ce77f4c317
http://arxiv.org/abs/2107.06140
http://arxiv.org/abs/2107.06140
Publikováno v:
Pattern Recognition. 72:407-418
Lifelong reinforcement learning provides a successful framework for agents to learn multiple consecutive tasks sequentially. Current methods, however, suffer from scalability issues when the agent has to solve a large number of tasks. In this paper,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8afb5978fe52512bb6d5edf1258af740
https://doi.org/10.1016/j.patcog.2017.07.031
https://doi.org/10.1016/j.patcog.2017.07.031
Publikováno v:
AAAI
In situations where explicit communication is limited, human collaborators act by learning to: (i) infer meaning behind their partner's actions, and (ii) convey private information about the state to their partner implicitly through actions. The firs
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::81a9afd652adfdd8ba28ea43022af2b9
http://arxiv.org/abs/1810.04444
http://arxiv.org/abs/1810.04444
Publikováno v:
IJCAI
Within the context of video games the notion of perfectly rational agents can be undesirable as it leads to uninteresting situations, where humans face tough adversarial decision makers. Current frameworks for stochastic games and reinforcement learn
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7f8528d671165c13688737f1acdec1a1
http://arxiv.org/abs/1802.03216
http://arxiv.org/abs/1802.03216
Autor:
Christopher Amato, Haitham Bou Ammar, Elizabeth Churchill, Erez Karpas, Takashi Kido, Mike Kuniavsky, W. F. Lawless, Francesca Rossi, Frans A. Oliehoek, Stephen Russell, Keiki Takadama, Siddharth Srivastava, Karl Tuyls, Philip Van Allen, K. Brent Venable, Peter Vrancx, Shiqi Zhang
Publikováno v:
AI Magazine, 39(4)
AI Magazine; Vol 39 No 4: Winter 2018; 29-35
AI Magazine; Vol 39 No 4: Winter 2018; 29-35
© 2018, Association for the Advancement of Artifcial Intelligence. All rights reserved. The Association for the Advancement of Artifcial Intelligence, in cooperation with Stanford University's Department of Computer Science, presented the 2018 Sprin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0860fb6a6a4b56f67338bf36536a2648
http://resolver.tudelft.nl/uuid:6e3c4531-44ef-4045-924e-09839b1d19be
http://resolver.tudelft.nl/uuid:6e3c4531-44ef-4045-924e-09839b1d19be