Neural Task Programming: Learning to Generalize Across Hierarchical Tasks

  title={Neural Task Programming: Learning to Generalize Across Hierarchical Tasks},
  author={Danfei Xu and Suraj Nair and Yuke Zhu and J. Gao and Animesh Garg and Li Fei-Fei and S. Savarese},
  journal={2018 IEEE International Conference on Robotics and Automation (ICRA)},
  • Danfei Xu, Suraj Nair, +4 authors S. Savarese
  • Published 2018
  • Computer Science
  • 2018 IEEE International Conference on Robotics and Automation (ICRA)
  • In this work, we propose a novel robot learning framework called Neural Task Programming (NTP), which bridges the idea of few-shot learning from demonstration and neural program induction. [...] Key Method These specifications are fed to a hierarchical neural program, where bottom-level programs are callable subroutines that interact with the environment. We validate our method in three robot manipulation tasks. NTP achieves strong generalization across sequential tasks that exhibit hierarchal and compositional…Expand Abstract
    86 Citations

    Figures and Topics from this paper

    Explore Further: Topics Discussed in This Paper

    Neural Task Graphs: Generalizing to Unseen Tasks From a Single Video Demonstration
    • 40
    • PDF
    Time Reversal as Self-Supervision
    • 8
    • PDF
    TACO: Learning Task Decomposition via Temporal Alignment for Control
    • 28
    • PDF
    Learning Compositional Neural Programs for Continuous Control
    • PDF
    One-Shot Hierarchical Imitation Learning of Compound Visuomotor Tasks
    • 25
    • PDF
    TRASS: Time Reversal as Self-Supervision
    Learning to Imagine Manipulation Goals for Robot Task Planning
    • 1
    • PDF
    Plan Arithmetic: Compositional Plan Vectors for Multi-Task Control
    • 2
    • PDF
    Transferable Task Execution from Pixels through Deep Planning Domain Learning
    • 3
    • PDF
    Meta Reinforcement Learning with Autonomous Inference of Subtask Dependencies
    • 3
    • Highly Influenced
    • PDF


    Modular Multitask Reinforcement Learning with Policy Sketches
    • 192
    • PDF
    Learning modular neural network policies for multi-task and multi-robot transfer
    • 184
    • Highly Influential
    • PDF
    Making Neural Programming Architectures Generalize via Recursion
    • 106
    • PDF
    One-Shot Imitation Learning
    • 340
    • PDF
    Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
    • 590
    • PDF
    Learning Sequences of Controllers for Complex Manipulation Tasks
    • 10
    Target-driven visual navigation in indoor scenes using deep reinforcement learning
    • 730
    • PDF
    A survey of robot learning from demonstration
    • 2,537
    • PDF
    A Deep Hierarchical Approach to Lifelong Learning in Minecraft
    • 214
    • PDF