Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Kim, Kuno"'
Many patterns in nature exhibit self-similarity: they can be compactly described via self-referential transformations. Said patterns commonly appear in natural and artificial objects, such as molecules, shorelines, galaxies and even images. In this w
Externí odkaz:
http://arxiv.org/abs/2204.07673
Learning policies that effectively utilize language instructions in complex, multi-task environments is an important problem in sequential decision-making. While it is possible to condition on the entire language instruction directly, such an approac
Externí odkaz:
http://arxiv.org/abs/2203.00054
We propose a new framework for Imitation Learning (IL) via density estimation of the expert's occupancy measure followed by Maximum Occupancy Entropy Reinforcement Learning (RL) using the density as a reward. Our approach maximizes a non-adversarial
Externí odkaz:
http://arxiv.org/abs/2010.09808
World models are self-supervised predictive models of how the world evolves. Humans learn world models by curiously exploring their environment, in the process acquiring compact abstractions of high bandwidth sensory inputs, the ability to plan acros
Externí odkaz:
http://arxiv.org/abs/2007.07853
Autor:
Gan, Chuang, Schwartz, Jeremy, Alter, Seth, Mrowca, Damian, Schrimpf, Martin, Traer, James, De Freitas, Julian, Kubilius, Jonas, Bhandwaldar, Abhishek, Haber, Nick, Sano, Megumi, Kim, Kuno, Wang, Elias, Lingelbach, Michael, Curtis, Aidan, Feigelis, Kevin, Bear, Daniel M., Gutfreund, Dan, Cox, David, Torralba, Antonio, DiCarlo, James J., Tenenbaum, Joshua B., McDermott, Josh H., Yamins, Daniel L. K.
We introduce ThreeDWorld (TDW), a platform for interactive multi-modal physical simulation. TDW enables simulation of high-fidelity sensory data and physical interactions between mobile agents and objects in rich 3D environments. Unique properties in
Externí odkaz:
http://arxiv.org/abs/2007.04954
We study the question of how to imitate tasks across domains with discrepancies such as embodiment, viewpoint, and dynamics mismatch. Many prior works require paired, aligned demonstrations and an additional RL step that requires environment interact
Externí odkaz:
http://arxiv.org/abs/1910.00105