Zobrazeno 1 - 10
of 21
pro vyhledávání: '"Daan Wierstra"'
Autor:
Oriol Vinyals, Theophane Weber, Daan Wierstra, S. M. Ali Eslami, Karol Gregor, Neil C. Rabinowitz, Demis Hassabis, Matthew Botvinick, Koray Kavukcuoglu, Frederic Besse, Fabio Viola, Helen Dean King, David P. Reichert, Andrei Rusu, Chloe Hillier, Ivo Danihelka, Lars Buesing, Dan Rosenbaum, Ari S. Morcos, Avraham Ruderman, Marta Garnelo, Danilo Jimenez Rezende
Publikováno v:
Science. 360:1204-1210
A scene-internalizing computer program To train a computer to “recognize” elements of a scene supplied by its visual sensors, computer scientists typically use millions of images painstakingly labeled by humans. Eslami et al. developed an artific
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::606988da59ae4eee1d6edc64410ed836
https://doi.org/10.1126/science.aar6170
https://doi.org/10.1126/science.aar6170
Autor:
Joel Veness, Andrei Rusu, Koray Kavukcuoglu, Alex Graves, Volodymyr Mnih, Daan Wierstra, Marc G. Bellemare, David Silver, Georg Ostrovski, Dharshan Kumaran, Andreas K. Fidjeland, Helen King, Stig Petersen, Martin Riedmiller, Demis Hassabis, Charles Beattie, Ioannis Antonoglou, Amir Sadik, Shane Legg
Publikováno v:
Nature. 518:529-533
The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successf
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b6e10fab190565e277a2761c6d168781
https://doi.org/10.1038/nature14236
https://doi.org/10.1038/nature14236
Autor:
Matthew Botvinick, David G. T. Barrett, Greg Wayne, Theophane Weber, Joseph Modayil, Timothy P. Lillicrap, Darshan Kumaran, Joel Z. Leibo, Peter W. Battaglia, Demis Hassabis, Christopher Summerfield, Nando de Freitas, Shakir Mohamed, Daan Wierstra, Adam Santoro, Danilo Jimenez Rezende, Neil C. Rabinowitz, Tom Schaul, Shane Legg
Publikováno v:
Behavioral and Brain Sciences
We agree with Lake and colleagues on their list of “key ingredients” for building human-like intelligence, including the idea that model-based reasoning is essential. However, we favor an approach that centers on one additional ingredient: autono
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8f66d844ed73f5cf2d78e8d50299b73a
https://doi.org/10.1017/s0140525x17000048
https://doi.org/10.1017/s0140525x17000048
Autor:
Dylan Banarse, Chrisantha Fernando, Malcolm Reynolds, Max Jaderberg, David Pfau, Marc Lanctot, Daan Wierstra, Frederic Besse
Publikováno v:
GECCO
In this work we introduce a differentiable version of the Compositional Pattern Producing Network, called the DPPN. Unlike a standard CPPN, the topology of a DPPN is evolved but the weights are learned. A Lamarckian algorithm, that combines evolution
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ccf1435d82e26177d492ef9a9ea17d94
https://doi.org/10.1145/2908812.2908890
https://doi.org/10.1145/2908812.2908890
Publikováno v:
Paladyn: Journal of Behavioral Robotics, Vol 1, Iss 1, Pp 14-24 (2010)
This paper discusses parameter-based exploration methods for reinforcement learning. Parameter-based methods perturb parameters of a general function approximator directly, rather than adding noise to the resulting actions. Parameter-based exploratio
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9bfe041169e61cbf6d4820d748f20dad
https://doaj.org/article/1d7d7f4ee4c14115b57878db43f50fab
https://doaj.org/article/1d7d7f4ee4c14115b57878db43f50fab
Publikováno v:
IROS
Tying suture knots is a time-consuming task performed frequently during Minimally Invasive Surgery (MIS). Automating this task could greatly reduce total surgery time for patients. Current solutions to this problem replay manually programmed trajecto
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::acfbbe562b1f7bb180be14852b48087e
https://doi.org/10.1163/156855308x360604
https://doi.org/10.1163/156855308x360604
Publikováno v:
Neural Computation. 19:757-779
In recent years, gradient-based LSTM recurrent neural networks (RNNs) solved many previously RNN-unlearnable tasks. Sometimes, however, gradient information is of little use for training RNNs, due to numerous local minima. For such cases, we present
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::78b97279ca11ec59bb4d96843e11f328
https://doi.org/10.1162/neco.2007.19.3.757
https://doi.org/10.1162/neco.2007.19.3.757
Publikováno v:
IEEE Congress on Evolutionary Computation
The principle of artificial curiosity directs active exploration towards the most informative or most interesting data. We show its usefulness for global black box optimization when data point evaluations are expensive. Gaussian process regression is
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e5dcd1e4c08a38992af34f2988bc7fed
Publikováno v:
GECCO
The family of natural evolution strategies (NES) offers a principled approach to real-valued evolutionary optimization by following the natural gradient of the expected fitness. Like the well-known CMA-ES, the most competitive algorithm in the field,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a1d9b1316bcf04dcdb2548d9a505ff1b
https://doi.org/10.1145/1830483.1830557
https://doi.org/10.1145/1830483.1830557
Publikováno v:
IEEE Congress on Evolutionary Computation
The problem of how to create NPC AI for videogames that believably imitates particular human players is addressed. Previous approaches to learning player behaviour is found to either not generalize well to new environments and noisy perceptions, or t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e599fc631c4bef42e7fc04c0ced101dc
https://mediatum.ub.tum.de/doc/1289110/document.pdf
https://mediatum.ub.tum.de/doc/1289110/document.pdf