Corpus ID: 198968158

To Learn or Not to Learn: Analyzing the Role of Learning for Navigation in Virtual Environments

@article{Kojima2019ToLO,
  title={To Learn or Not to Learn: Analyzing the Role of Learning for Navigation in Virtual Environments},
  author={Noriyuki Kojima and Jia Deng},
  journal={ArXiv},
  year={2019},
  volume={abs/1907.11770}
}
  • Noriyuki Kojima, Jia Deng
  • Published 2019
  • Computer Science
  • ArXiv
  • In this paper we compare learning-based methods and classical methods for navigation in virtual environments. [...] Key Method We perform detailed analysis to study the strengths and weaknesses of learned agents and classical agents, as well as how characteristics of the virtual environment impact navigation performance. Our results show that learned agents have inferior collision avoidance and memory management, but are superior in handling ambiguity and noise. These results can inform future design of…Expand Abstract

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.
    SHOWING 1-5 OF 5 CITATIONS

    Audio-Visual Embodied Navigation

    VIEW 5 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Habitat: A Platform for Embodied AI Research

    VIEW 8 EXCERPTS
    CITES RESULTS, METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Habitat Sim Generic Dataset Support Habitat API Habitat Platform

    VIEW 8 EXCERPTS
    CITES RESULTS, BACKGROUND & METHODS
    HIGHLY INFLUENCED

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 41 REFERENCES

    Cognitive Mapping and Planning for Visual Navigation

    VIEW 8 EXCERPTS
    HIGHLY INFLUENTIAL

    Neural SLAM

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning to Act by Predicting the Future

    VIEW 3 EXCERPTS

    Gibson Env: Real-World Perception for Embodied Agents

    VIEW 1 EXCERPT