Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning
© 2020 National Academy of Sciences. All rights reserved. Many animals, and an increasing number of artificial agents, display sophisticated capabilities to perceive and manipulate objects. But human beings remain distinctive in their capacity for flexible, creative tool use-using objects in new way...
Main Authors: | Allen, Kelsey R, Smith, Kevin A, Tenenbaum, Joshua B |
---|---|
Other Authors: | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences |
Format: | Article |
Language: | English |
Published: |
Proceedings of the National Academy of Sciences
2021
|
Online Access: | https://hdl.handle.net/1721.1/134068 |
Similar Items
-
Differentiable Physics and Stable Modes for Tool-Use and Manipulation Planning
by: Toussaint, Marc A, et al.
Published: (2020) -
End-to-end differentiable physics for learning and control
by: Smith, Kevin A, et al.
Published: (2020) -
Reasoning about physical interactions with object-oriented prediction and planning
by: Janner, Michael, et al.
Published: (2020) -
Lifelong learning of cognitive styles for physical problem-solving: The effect of embodied experience
by: Allen, Kelsey R., et al.
Published: (2023) -
Flexible neural representation for physics prediction
by: Mrowca, Damian, et al.
Published: (2020)