Corpus ID: 220935727

A Causal Lens for Peeking into Black Box Predictive Models: Predictive Model Interpretation via Causal Attribution

@article{Khademi2020ACL,
  title={A Causal Lens for Peeking into Black Box Predictive Models: Predictive Model Interpretation via Causal Attribution},
  author={Aria Khademi and Vasant G Honavar},
  journal={ArXiv},
  year={2020},
  volume={abs/2008.00357}
}
  • Aria Khademi, Vasant G Honavar
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • With the increasing adoption of predictive models trained using machine learning across a wide range of high-stakes applications, e.g., health care, security, criminal justice, finance, and education, there is a growing need for effective techniques for explaining such models and their predictions. We aim to address this problem in settings where the predictive model is a black box; That is, we can only observe the response of the model to various inputs, but have no knowledge about the… CONTINUE READING
    Commuting Network Spillovers and COVID-19 Deaths Across US Counties

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 116 REFERENCES
    Identification of Causal Effects Using Instrumental Variables
    • 3,087
    • PDF
    The central role of the propensity score in observational studies for causal effects
    • 21,188
    • PDF
    Long Short-Term Memory
    • 30,901
    • PDF
    "Why Should I Trust You?": Explaining the Predictions of Any Classifier
    • 3,225
    • PDF
    Marginal Structural Models and Causal Inference in Epidemiology
    • 3,443
    • PDF
    A Unified Approach to Interpreting Model Predictions
    • 1,370
    • PDF
    Representation Learning: A Review and New Perspectives
    • 6,241
    • PDF
    Greedy function approximation: A gradient boosting machine.
    • 5,924
    • Highly Influential
    • PDF
    Targeted Learning: Causal Inference for Observational and Experimental Data
    • 542
    • PDF
    Visualizing and Understanding Convolutional Networks
    • 8,301
    • PDF