Corpus ID: 59413829

Personalization and Optimization of Decision Parameters via Heterogenous Causal Effects

  title={Personalization and Optimization of Decision Parameters via Heterogenous Causal Effects},
  author={Y. Tu and Kinjal Basu and Jinyun Yan and B. Tiwana and Shaunak Chatterjee},
  • Y. Tu, Kinjal Basu, +2 authors Shaunak Chatterjee
  • Published 2019
  • Computer Science, Mathematics
  • ArXiv
  • Randomized experimentation (also known as A/B testing or bucket testing) is very commonly used in the internet industry to measure the effect of a new treatment. Often, the decision on the basis of such A/B testing is to ramp the treatment variant that did best for the entire population. However, the effect of any given treatment varies across experimental units, and choosing a single variant to ramp to the whole population can be quite suboptimal. In this work, we propose a method which… CONTINUE READING


    Publications referenced by this paper.
    Heterogeneous Treatment Effects in Digital Experimentation
    • 16
    Estimating treatment effect heterogeneity in randomized program evaluation
    • 271
    • PDF
    Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
    • 697
    • Highly Influential
    • PDF
    Estimating causal effects of treatments in randomized and nonrandomized studies.
    • 6,213
    • Highly Influential
    • PDF
    Recursive partitioning for heterogeneous causal effects
    • 479
    • Highly Influential
    • PDF
    Subgroup identification from randomized clinical trial data.
    • 324
    • PDF
    The central role of the propensity score in observational studies for causal effects
    • 21,200
    • PDF