Zobrazeno 1 - 10
of 23
pro vyhledávání: '"Tom Erez"'
Autor:
Yuval Tassa, Vu Pham, Arun Ahuja, Leonard Hasenclever, Saran Tunyasuvunakool, Nicolas Heess, Josh Merel, Tom Erez, Greg Wayne
We address the longstanding challenge of producing flexible, realistic humanoid character controllers that can perform diverse whole-body tasks involving object interactions. This challenge is central to a variety of fields, from graphics and animati
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7dcf1b4b4a50661167c302565e0c4d47
Autor:
Ziyu Wang, Nando de Freitas, Saran Tunyasuvunakool, Raia Hadsell, Andrei Rusu, Serkan Cabi, Yuke Zhu, Josh Merel, Nicolas Heess, Tom Erez, János Kramár
Publikováno v:
Robotics: Science and Systems
We propose a model-free deep reinforcement learning method that leverages a small amount of demonstration data to assist a reinforcement learning agent. We apply this approach to robotic manipulation tasks and train end-to-end visuomotor policies tha
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e223b3355ed5af0a9400def814b8a551
http://arxiv.org/abs/1802.09564
http://arxiv.org/abs/1802.09564
Autor:
Tom Erez, Aviv Rubinstein
Publikováno v:
2016 IEEE International Conference on the Science of Electrical Engineering (ICSEE).
We present a novel turnkey system for autonomous tunnel mapping called LiTANK — LiDAR TANK. The system depends entirely on LiDAR sensor readings, without the use of other sensors such as GPS or odometry. Using point cloud registration to perform re
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2fdb607c27a24db876414587a9506591
https://doi.org/10.1109/icsee.2016.7806111
https://doi.org/10.1109/icsee.2016.7806111
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 25:952-957
This paper presents a generative model of eye-hand coordination. We use numerical optimization to solve for the joint behavior of an eye and two hands, deriving a predicted motion pattern from first principles, without imposing heuristics. We model t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::31a97bdeeed5f2bfc152b54b3ffd684f
https://doi.org/10.1609/aaai.v25i1.8007
https://doi.org/10.1609/aaai.v25i1.8007
Publikováno v:
IFAC Proceedings Volumes. 44:4707-4713
We introduce an algorithm that generates an optimal controller for stochastic nonlinear problems with a periodic solution, e.g. locomotion. Uniquely, the quantity we approximate is neither the Value nor Policy functions, but rather the stationary sta
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2a02e28e79d10172db27910aebbde616
https://doi.org/10.3182/20110828-6-it-1002.00903
https://doi.org/10.3182/20110828-6-it-1002.00903
Autor:
Lev Muchnik, Lea Tsaban, Royi Itzhack, Sorin Solomon, Tom Erez, Jacob Goldenberg, Yoram Louzoun
Publikováno v:
Physica A: Statistical Mechanics and its Applications. 389:5308-5318
The generation mechanisms of real world networks have been described using multiple models. The mathematical features of these models are usually extrapolated from statistical properties of a snapshot of these networks. We here propose an alternative
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::91063aee131eef67150cba8ebfc4c70d
https://doi.org/10.1016/j.physa.2010.07.011
https://doi.org/10.1016/j.physa.2010.07.011
Autor:
Yuval Tassa, Tom Erez
Publikováno v:
IEEE Transactions on Neural Networks. 18:1031-1041
In this paper, we present an empirical study of iterative least squares minimization of the Hamilton-Jacobi-Bellman (HJB) residual with a neural network (NN) approximation of the value function. Although the nonlinearities in the optimal control prob
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6f39d0a7b7429e67847624fe7b646291
https://doi.org/10.1109/tnn.2007.899249
https://doi.org/10.1109/tnn.2007.899249
Publikováno v:
ICRA
There is growing need for software tools that can accurately simulate the complex dynamics of modern robots. While a number of candidates exist, the field is fragmented. It is difficult to select the best tool for a given project, or to predict how m
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::997678c1dd9b03fcc1555d02d8d4e2e7
https://doi.org/10.1109/icra.2015.7139807
https://doi.org/10.1109/icra.2015.7139807
Publikováno v:
IROS
We describe a computationally-expensive but very accurate approach to state estimation, which fuses any available sensor data with physical consistency priors. This is done by combining the advantages of recursive estimation and fixedlag smoothing: a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ed50d0fca9bbd3be4ef3dc01092102e8
https://doi.org/10.1109/iros.2014.6942777
https://doi.org/10.1109/iros.2014.6942777
Publikováno v:
ICRA
Dexterous hand manipulation is one of the most complex types of biological movement, and has proven very difficult to replicate in robots. The usual approaches to robotic control – following pre-defined trajectories or planning online with reduced
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ebb9f6a89a14020d00d2f9945af924ca
https://doi.org/10.1109/icra.2014.6907864
https://doi.org/10.1109/icra.2014.6907864