Authors
Chelsea Finn, Ian Goodfellow, Sergey Levine
Publication date
2016/5/23
Journal
arXiv preprint arXiv:1605.07157
Description
Abstract: A core challenge for an agent learning to interact with the world is to predict how its
actions affect objects in its environment. Many existing methods for learning the dynamics of
physical interactions require labeled object information. However, to scale real-world
interaction learning to a variety of scenes and objects, acquiring labeled data becomes
increasingly impractical. To learn about physical object motion without labels, we develop
an action-conditioned video prediction model that explicitly models pixel motion, by ...
Total citations
20161
Scholar articles
C Finn, I Goodfellow, S Levine - arXiv preprint arXiv:1605.07157, 2016