Stochastic prediction of multi-agent interactions from partial observations
We present a method that learns to integrate temporal information, from a learned dynamics model, with ambiguous visual information, from a learned vision model, in the context of interacting agents. Our method is based on a graph-structured variational recurrent neural network (Graph-VRNN), which i...
Main Authors: | Sun, Chen, Karlsson, Per, Wu, Jiajun, Tenenbaum, Joshua B, Murphy, Kevin P |
---|---|
Other Authors: | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
Format: | Article |
Language: | English |
Published: |
International Conference on Learning Representations
2020
|
Online Access: | https://hdl.handle.net/1721.1/126593 |
Similar Items
-
Propagation networks for model-based control under partial observation
by: Li, Yunzhu, et al.
Published: (2020) -
Stochastic control of partially observable systems /
by: 238148 Bensoussan, Alain
Published: (2004) -
Learning for multi-robot cooperation in partially observable stochastic environments with macro-actions
by: Amato, Christopher, et al.
Published: (2018) -
Reasoning about physical interactions with object-oriented prediction and planning
by: Janner, Michael, et al.
Published: (2020) -
DensePhysNet: Learning Dense Physical Object Representations Via Multi-Step Dynamic Interactions
by: Xu, Zhenjia, et al.
Published: (2021)