Learning to Learn How to Learn: Self-Adaptive Visual Navigation Using Meta-Learning

@article{Wortsman2018LearningTL,
  title={Learning to Learn How to Learn: Self-Adaptive Visual Navigation Using Meta-Learning},
  author={Mitchell Wortsman and Kiana Ehsani and Mohammad Rastegari and Ali Farhadi and Roozbeh Mottaghi},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2018},
  pages={6743-6752}
}
Highlight Information
Learning is an inherently continuous phenomenon. [...] Key Method Our solution is a meta-reinforcement learning approach where an agent learns a self-supervised interaction loss that encourages effective navigation. Our experiments, performed in the AI2-THOR framework, show major improvements in both success rate and SPL for visual navigation in novel scenes. Our code and data are available at: https://github.com/allenai/savn.Expand Abstract
24
Twitter Mentions

Citations

Publications citing this paper.
SHOWING 1-10 OF 12 CITATIONS

Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation

VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Visual Object Search by Learning Spatial Context

VIEW 3 EXCERPTS
CITES METHODS & BACKGROUND

Audio-Visual Embodied Navigation

VIEW 1 EXCERPT
CITES METHODS

MAME : Model-Agnostic Meta-Exploration

VIEW 1 EXCERPT
CITES BACKGROUND

Meta Adaptation using Importance Weighted Demonstrations

VIEW 1 EXCERPT
CITES BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-10 OF 54 REFERENCES

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

Asynchronous Methods for Deep Reinforcement Learning

VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning

VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Learning and Planning with a Semantic Model

VIEW 1 EXCERPT

Learning to Adapt: Meta-Learning for Model-Based Control

VIEW 1 EXCERPT

Learning to Explore with Meta-Policy Gradient

VIEW 1 EXCERPT