Corpus ID: 67788344

From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following

@article{Fu2019FromLT,
  title={From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following},
  author={Justin Fu and A. Balan and S. Levine and S. Guadarrama},
  journal={ArXiv},
  year={2019},
  volume={abs/1902.07742}
}
  • Justin Fu, A. Balan, +1 author S. Guadarrama
  • Published 2019
  • Mathematics, Computer Science
  • ArXiv
  • Reinforcement learning is a promising framework for solving control problems, but its use in practical situations is hampered by the fact that reward functions are often difficult to engineer. [...] Key Method We propose language-conditioned reward learning (LC-RL), which grounds language commands as a reward function represented by a deep neural network. We demonstrate that our model learns rewards that transfer to novel tasks and environments on realistic, high-dimensional visual environments with natural…Expand Abstract
    Inverse Reinforcement Learning with Natural Language Goals
    Language as a Cognitive Tool to Imagine Goals in Curiosity-Driven Exploration
    5
    Learning a Prior over Intent via Meta-Inverse Reinforcement Learning
    20
    Language as an Abstraction for Hierarchical Deep Reinforcement Learning
    19
    Inverse Reinforcement Learning in Contextual MDPs

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 30 REFERENCES
    Learning to Follow Language Instructions with Adversarial Reward Induction
    22
    Deep Reinforcement Learning from Human Preferences
    302
    Grounding English Commands to Reward Functions
    47
    Reward Learning from Narrated Demonstrations
    13
    Learning Robust Rewards with Adversarial Inverse Reinforcement Learning
    175
    Human-level control through deep reinforcement learning
    9336
    Reinforcement Learning for Mapping Instructions to Actions
    221