# Sources of Gain: Decomposing Performance in Conditional Average Dose Response Estimation

arXiv: https://arxiv.org/abs/2406.08206 · HTML: https://mlsanity.com/paper/2406.08206/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 79 references that are in the map.
- First posted on arXiv 12 Jun 2024.

## Builds on (most cited first, 23 of 23)

- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Estimation and Inference of Heterogeneous Treatment Effects using Random Forests](https://mlsanity.com/paper/1510.04342/) (2015)
- [Generalizing from a Few Examples: A Survey on Few-Shot Learning](https://mlsanity.com/paper/1904.05046/) (2019)
- [Causal Effect Inference with Deep Latent-Variable Models](https://mlsanity.com/paper/1705.08821/) (2017)
- [Learning Representations for Counterfactual Inference](https://mlsanity.com/paper/1605.03661/) (2016)
- [Adapting Neural Networks for the Estimation of Treatment Effects](https://mlsanity.com/paper/1906.02120/) (2019)
- [OpenOOD: Benchmarking Generalized Out-of-Distribution Detection](https://mlsanity.com/paper/2210.07242/) (2022)
- [Learning Counterfactual Representations for Estimating Individual Dose-Response Curves](https://mlsanity.com/paper/1902.00981/) (2019)
- [OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalization](https://mlsanity.com/paper/2106.03721/) (2021)
- [Estimating the Effects of Continuous-valued Interventions using Generative Adversarial Networks](https://mlsanity.com/paper/2002.12326/) (2020)
- [Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effects](https://mlsanity.com/paper/2001.07426/) (2020)
- [VCNet and Functional Targeted Regularization For Learning Causal Effects of Continuous Treatments](https://mlsanity.com/paper/2103.07861/) (2021)
- [k-Nearest Neighbour Classifiers: 2nd Edition (with Python examples)](https://mlsanity.com/paper/2004.04523/) (2020)
- [Kernel Methods for Causal Functions: Dose, Heterogeneous, and Incremental Response Curves](https://mlsanity.com/paper/2010.04855/) (2020)
- [In Search of Insights, Not Magic Bullets: Towards Demystification of the Model Selection Dilemma in Heterogeneous Treatment Effect Estimation](https://mlsanity.com/paper/2302.02923/) (2023)
- [A comparison of methods for model selection when estimating individual treatment effects](https://mlsanity.com/paper/1804.05146/) (2018)
- [Exploring Transformer Backbones for Heterogeneous Treatment Effect Estimation](https://mlsanity.com/paper/2202.01336/) (2022)
- [Reliable Off-Policy Learning for Dosage Combinations](https://mlsanity.com/paper/2305.19742/) (2023)
- [Entropy Balancing for Continuous Treatments](https://mlsanity.com/paper/2001.06281/) (2020)
- [Machine learning on small size samples: A synthetic knowledge synthesis](https://mlsanity.com/paper/2103.01002/) (2021)
- [Adversarially Balanced Representation for Continuous Treatment Effect Estimation](https://mlsanity.com/paper/2312.10570/) (2023)
- [Continuous Treatment Effect Estimation Using Gradient Interpolation and Kernel Smoothing](https://mlsanity.com/paper/2401.15447/) (2024)
- [Contrastive Balancing Representation Learning for Heterogeneous Dose-Response Curves Estimation](https://mlsanity.com/paper/2403.14232/) (2024)

## Built on by (most cited first, 2 of 2)

- [Uplift modeling with continuous treatments: A predict-then-optimize approach](https://mlsanity.com/paper/2412.09232/) (2024)
- [Causal Machine Learning for Patient-Level Intraoperative Opioid Dose Prediction from Electronic Health Records](https://mlsanity.com/paper/2508.09059/) (2025)

## Abstract (arXiv)

Estimating conditional average dose responses (CADR) is an important but challenging problem. Estimators must correctly model the potentially complex relationships between covariates, interventions, doses, and outcomes. In recent years, the machine learning community has shown great interest in developing tailored CADR estimators that target specific challenges. Their performance is typically evaluated against other methods on (semi-) synthetic benchmark datasets. Our paper analyses this practice and shows that using popular benchmark datasets without further analysis is insufficient to judge model performance. Established benchmarks entail multiple challenges, whose impacts must be disentangled. Therefore, we propose a novel decomposition scheme that allows the evaluation of the impact of five distinct components contributing to CADR estimator performance. We apply this scheme to eight popular CADR estimators on four widely-used benchmark datasets, running nearly 1,500 individual experiments. Our results reveal that most established benchmarks are challenging for reasons different from their creators' claims. Notably, confounding, the key challenge tackled by most estimators, is not an issue in any of the considered datasets. We discuss the major implications of our findings and present directions for future research.