Corpus ID: 211082703

Efficient Policy Learning from Surrogate-Loss Classification Reductions

@article{Bennett2020EfficientPL,
  title={Efficient Policy Learning from Surrogate-Loss Classification Reductions},
  author={Andrew Bennett and Nathan Kallus},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.05153}
}
  • Andrew Bennett, Nathan Kallus
  • Published 2020
  • Mathematics, Computer Science, Economics
  • ArXiv
  • Recent work on policy learning from observational data has highlighted the importance of efficient policy evaluation and has proposed reductions to weighted (cost-sensitive) classification. But, efficient policy evaluation need not yield efficient estimation of policy parameters. We consider the estimation problem given by a weighted surrogate-loss classification reduction of policy learning with any score function, either direct, inverse-propensity weighted, or doubly robust. We show that… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 30 REFERENCES

    Entropy Learning for Dynamic Treatment Regimes.

    VIEW 12 EXCERPTS
    HIGHLY INFLUENTIAL

    Offline Multi-Action Policy Learning: Generalization and Optimization

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Efficient Policy Learning

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Residual Weighted Learning for Estimating Individualized Treatment Rules.

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Estimating Individualized Treatment Rules Using Outcome Weighted Learning.

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Doubly Robust Policy Evaluation and Learning

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    The offset tree for learning with partial labels

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Asymptotic statistics

    • A. W. Van der Vaart
    • 2000
    VIEW 2 EXCERPTS
    HIGHLY INFLUENTIAL