Corpus ID: 7624474

A Deep Compositional Framework for Human-like Language Acquisition in Virtual Environment

@article{Yu2017ADC,
  title={A Deep Compositional Framework for Human-like Language Acquisition in Virtual Environment},
  author={Haonan Yu and H. Zhang and W. Xu},
  journal={ArXiv},
  year={2017},
  volume={abs/1703.09831}
}
  • Haonan Yu, H. Zhang, W. Xu
  • Published 2017
  • Computer Science
  • ArXiv
  • We tackle a task where an agent learns to navigate in a 2D maze-like environment called XWORLD. [...] Key Method Our deep framework for the agent is trained end to end: it learns simultaneously the visual representations of the environment, the syntax and semantics of the language, and the action module that outputs actions. The zero-shot learning capability of our framework results from its compositionality and modularity with parameter tying. We visualize the intermediate outputs of the framework…Expand Abstract

    Figures, Tables, and Topics from this paper.

    Explore Further: Topics Discussed in This Paper

    Grounding Language in Play
    1
    Grounded Language Learning in a Simulated 3D World
    137
    Deep Reinforcement Learning for Instruction Following Visual Navigation in 3D Maze-Like Environments
    2
    Embodied Question Answering
    215
    Embodied Question Answering
    10
    A new dataset and model for learning to understand navigational instructions
    1
    Hierarchical Reinforcement Learning for Zero-shot Generalization with Subtask Dependencies
    10

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 35 REFERENCES
    Building Machines That Learn and Think Like People
    905
    Virtual Embodiment: A Scalable Long-Term Strategy for Artificial Intelligence Research
    28
    Robot Language Learning, Generation, and Comprehension
    11
    ViZDoom: A Doom-based AI research platform for visual reinforcement learning
    392
    Reinforcement Learning with Unsupervised Auxiliary Tasks
    605
    Human-level control through deep reinforcement learning
    9494
    Neural Module Networks
    474