# De-paradox Tree: Breaking Down Simpson's Paradox via A Kernel-Based Partition Algorithm

arXiv: https://arxiv.org/abs/2603.02174 · HTML: https://mlsanity.com/paper/2603.02174/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 111 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Builds on (most cited first, 28 of 28)

- [Estimation and Inference of Heterogeneous Treatment Effects using Random Forests](https://mlsanity.com/paper/1510.04342/) (2015)
- [Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning](https://mlsanity.com/paper/1706.03461/) (2017)
- [Kernel Mean Embedding of Distributions: A Review and Beyond](https://mlsanity.com/paper/1605.09522/) (2016)
- [Recursive Partitioning for Heterogeneous Causal Effects](https://mlsanity.com/paper/1504.01132/) (2015)
- [Doubly Robust Policy Evaluation and Optimization](https://mlsanity.com/paper/1503.02834/) (2015)
- [Orthogonal Statistical Learning](https://mlsanity.com/paper/1901.09036/) (2019)
- [CausalML: Python Package for Causal Machine Learning](https://mlsanity.com/paper/2002.11631/) (2020)
- [Offline Multi-Action Policy Learning: Generalization and Optimization](https://mlsanity.com/paper/1810.04778/) (2018)
- [Orthogonal Random Forest for Causal Inference](https://mlsanity.com/paper/1806.03467/) (2018)
- [Uplift Modeling with Multiple Treatments and General Response Types](https://mlsanity.com/paper/1705.08492/) (2017)
- [Covariate Balancing Propensity Score by Tailored Loss Functions](https://mlsanity.com/paper/1601.05890/) (2016)
- [Machine Learning Estimation of Heterogeneous Treatment Effects with Instruments](https://mlsanity.com/paper/1905.10176/) (2019)
- [Orthogonal Machine Learning: Power and Limitations](https://mlsanity.com/paper/1711.00342/) (2017)
- [Are All Spurious Features in Natural Language Alike? An Analysis through a Causal Lens](https://mlsanity.com/paper/2210.14011/) (2022)
- [Model-based Recursive Partitioning for Subgroup Analyses](https://mlsanity.com/paper/1605.04078/) (2016)
- [The Balancing Act in Causal Inference](https://mlsanity.com/paper/2110.14831/) (2021)
- [Regularized calibrated estimation of propensity scores with model misspecification and high-dimensional data](https://mlsanity.com/paper/1710.08074/) (2017)
- [Interpretable Subgroup Discovery in Treatment Effect Estimation with Application to Opioid Prescribing Guidelines](https://mlsanity.com/paper/1905.03297/) (2019)
- [Using Decision Lists to Construct Interpretable and Parsimonious Treatment Regimes](https://mlsanity.com/paper/1504.07715/) (2015)
- [Simpson's paradox in Covid-19 case fatality rates: a mediation analysis of age-related causal effects](https://mlsanity.com/paper/2005.07180/) (2020)
- [Kernel Balancing: A flexible non-parametric weighting procedure for estimating causal effects](https://mlsanity.com/paper/1605.00155/) (2016)
- [Using Simpson's Paradox to Discover Interesting Patterns in Behavioral Data](https://mlsanity.com/paper/1805.03094/) (2018)
- [Learning Triggers for Heterogeneous Treatment Effects](https://mlsanity.com/paper/1902.00087/) (2019)
- [VAINE: Visualization and AI for Natural Experiments](https://mlsanity.com/paper/2109.04348/) (2021)
- [Can you Trust the Trend: Discovering Simpson's Paradoxes in Social Data](https://mlsanity.com/paper/1801.04385/) (2018)
- [Computational Social Scientist Beware: Simpson's Paradox in Behavioral Data](https://mlsanity.com/paper/1710.08615/) (2017)
- [Learning and Testing Sub-groups with Heterogeneous Treatment Effects:A Sequence of Two Studies](https://mlsanity.com/paper/2006.11628/) (2020)
- [DoGR: Disaggregated Gaussian Regression for Reproducible Analysis of Heterogeneous Data](https://mlsanity.com/paper/2108.13581/) (2021)

## Built on by (most cited first, 1 of 1)

- [PROXIMA: A Reliability Scoring Framework for Proxy Metrics in Online Controlled Experiments](https://mlsanity.com/paper/2604.14352/) (2026)

## Abstract (arXiv)

Real-world observational datasets and machine learning have revolutionized data-driven decision-making, yet many models rely on empirical associations that may be misleading due to confounding and subgroup heterogeneity. Simpson's paradox exemplifies this challenge, where aggregated and subgroup-level associations contradict each other, leading to misleading conclusions. Existing methods provide limited support for detecting and interpreting such paradoxical associations, especially for practitioners without deep causal expertise. We introduce De-paradox Tree, an interpretable algorithm designed to uncover hidden subgroup patterns behind paradoxical associations under assumed causal structures involving confounders and effect heterogeneity. It employs novel split criteria and balancing-based procedures to adjust for confounders and homogenize heterogeneous effects through recursive partitioning. Compared to state-of-the-art methods, De-paradox Tree builds simpler, more interpretable trees, selects relevant covariates, and identifies nested opposite effects while ensuring robust estimation of causal effects when causally admissible variables are provided. Our approach addresses the limitations of traditional causal inference and machine learning methods by introducing an interpretable framework that supports non-expert practitioners while explicitly acknowledging causal assumptions and scope limitations, enabling more reliable and informed decision-making in complex observational data environments.