Explainable Neural Computation via Stack Neural Module Networks

@article{Hu2018ExplainableNC,
  title={Explainable Neural Computation via Stack Neural Module Networks},
  author={Ronghang Hu and Jacob Andreas and Trevor Darrell and Kate Saenko},
  journal={ArXiv},
  year={2018},
  volume={abs/1807.08556}
}
  • Ronghang Hu, Jacob Andreas, +1 author Kate Saenko
  • Published 2018
  • Computer Science
  • ArXiv
  • In complex inferential tasks like question answering, machine learning models must confront two challenges: the need to implement a compositional reasoning process, and, in many applications, the need for this reasoning process to be interpretable to assist users in both development and prediction. Existing models designed to produce interpretable traces of their decision-making process typically require these traces to be supervised at training time. In this paper, we present a novel neural… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 59 CITATIONS

    Interpretable Visual Question Answering by Reasoning on Dependency Trees

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    INTERPRETABLE VISUAL QUESTION ANSWERING

    • 2018
    VIEW 3 EXCERPTS
    CITES RESULTS, METHODS & BACKGROUND

    Probabilistic Neural-symbolic Models for Interpretable Visual Question Answering

    VIEW 8 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Meta Module Network for Compositional Visual Reasoning

    VIEW 3 EXCERPTS
    CITES BACKGROUND

    Learning from Language

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Learning to Assemble Neural Module Tree Networks for Visual Grounding

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Perceptual Visual Reasoning with Knowledge Propagation

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    Structure Learning for Neural Module Networks

    VIEW 2 EXCERPTS
    CITES METHODS

    SYSTEMATIC GENERALIZATION: WHAT IS REQUIRED

    • 2018
    VIEW 2 EXCERPTS
    CITES BACKGROUND & RESULTS

    FILTER CITATIONS BY YEAR

    2018
    2020

    CITATION STATISTICS

    • 9 Highly Influenced Citations

    • Averaged 19 Citations per year from 2018 through 2020

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 42 REFERENCES

    Compositional Attention Networks for Machine Reasoning

    VIEW 10 EXCERPTS
    HIGHLY INFLUENTIAL

    A simple neural network module for relational reasoning

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning to Reason: End-to-End Module Networks for Visual Question Answering

    VIEW 10 EXCERPTS

    Inferring and Executing Programs for Visual Reasoning

    VIEW 10 EXCERPTS
    HIGHLY INFLUENTIAL

    CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning

    VIEW 3 EXCERPTS

    Neural Module Networks

    VIEW 4 EXCERPTS

    Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning

    VIEW 10 EXCERPTS
    HIGHLY INFLUENTIAL