Method uses pseudo-samples to improve RCT data in ride-hailing pricing studies.
problem Small and biased RCT data leads to significant bias when generalizing to broader user base.
method Pseudo-sample matching to expand and match RCT data with observational data.
result 0.41% improvement in profit through pseudo-sample matching.
CVIB uses information theory to learn counterfactuals from MNAR data without RCTs.
problem Debiasing learning from missing-not-at-random (MNAR) data in recommendation systems.
method CVIB, a variational information bottleneck, separates task-aware mutual information into factual and counterfactual parts.
result CVIB significantly enhances both shallow and deep models in recommendation systems.
The paper benchmarks OS with RCTs, accounting for right-censoring.
problem Benchmarking observational studies with experimental data under censoring.
method Two cases: independent and dependent censoring. Censoring-doubly-robust signal for CATE.
result Effectiveness of censoring-aware tests verified via experiments and real data.
A-TMLE estimates ATE from RCT and RWD, achieving super-efficiency.
problem Estimating ATE from RCT and RWD data.
method Adaptive-TMLE framework for decomposing and estimating ATE.
result A-TMLE is root-n consistent and asymptotically normal, achieving super-efficiency.
SDD improves DD for estimating treatment effects by adjusting for confounding.
problem Estimating treatment effects in observational studies with confounding.
method Synthesized Difference in Differences (SDD) using RCT data to infer correct slopes.
result SDD achieves state-of-the-art performance across synthetic and real datasets.
Method estimates CATE using RCT data to handle hidden confounders.
problem Estimating CATE in the presence of hidden confounders.
method Pseudo-confounder generator and CATE model alignment.
result Method reduces bias in CATE estimation.
Observed events in recommendation are consequence of the decisions made by a policy, thus they are usually selectively labeled, namely the data are Missing Not At Random (MNAR), which often causes large bias to the estimate of true outcomes risk. A general approach to correct MNAR bias is performing small Randomized Co…
Bayesian PROCOVA uses AI to adjust for covariates in RCTs.
problem Unbiased and precise treatment effect inferences from RCTs.
method Generative AI constructs digital twins for covariate adjustment, using an additive mixture prior.
result Efficiency gains in smaller RCTs compared to frequentist methods.
Paper proposes a novel method to assess treatment effect estimators using cross-validation.
problem Lack of ground truth to objectively assess treatment effect estimators in RCTs.
method Cross-validation-like methodology combining noisy difference-of-means estimate and aggregation across RCTs.
result Aggressive downweighting or truncation of large values reduces variance and improves treatment effect estimation.
Our paper improves uplift model evaluation on randomized controlled trials (RCT) data.
problem Variance in uplift evaluation metrics makes their signals arbitrary and unreliable.
method Theoretical analysis and statistical adjustment of the outcome to reduce variance.
result Variance reduction methods improve uplift evaluation metrics on RCT data.
A new method improves treatment effect inferences in RCTs by adjusting for covariates and heteroskedasticity.
problem Improving treatment effect inferences in RCTs with efficient and powerful methods.
method Weighted Prognostic Covariate Adjustment Method (Weighted PROCOVA) for heteroskedasticity.
result The method reduces variance, maintains Type I error rate, and increases test power for treatment effect.
Machine learning boosts RCT efficiency by controlling type I error and improving statistical power.
problem Improving statistical efficiency in RCTs with complex covariate adjustments.
method Machine learning-assisted adjustment under Rosenbaum's framework for exact tests.
result The proposed method robustly controls type I error and significantly boosts statistical efficiency.
New estimator improves policy evaluation in resource allocation RCTs.
problem Difficulty in evaluating policies optimizing limited resource allocation through RCTs.
method Proposes a novel estimator involving retrospective reshuffling of participants across experimental arms.
result The new estimator provides more accurate policy evaluations than common methods.
New method optimizes experimental design for specific applications.
problem Inability to adapt causal inference methods to specific downstream applications.
method Task-specific experimental design and sampling strategies.
result Significantly reduces data requirements for achieving RCT performance.
Participants enrolled into randomized controlled trials (RCTs) often do not reflect real-world populations. Previous research in how best to translate RCT results to target populations has focused on weighting RCT data to look like the target data. Simulation work, however, has suggested that an outcome model approach …
Transformer learns shipping costs more accurately than traditional methods.
problem Inaccurate shipping cost estimates lead to poor financial decisions.
method Proposes Rate Card Transformer (RCT) using self-attention to encode shipping information.
result Cost predictions made by RCT have 28.82% less error compared to GBDT models.
Prognostic scores improve logistic regression analysis in RCTs with binary outcomes.
problem Non-collapsibility in logistic regression analysis of RCTs with binary endpoints.
method Prognostic score adjustment using AI predictions to address non-collapsibility.
result Prognostic score adjustment increases power or reduces sample size for estimating conditional odds ratios.
New study finds targeting based on treatment effects outperforms risk-based targeting in social interventions.
problem Lack of accurate treatment effect estimates for machine learning-based targeting in social domains.
method Empirical assessment of targeting strategies using data from 5 real-world RCTs in various domains.
result Treatment effect-based targeting outperforms risk-based targeting, even with biased estimates.
In this paper we present tools for applied researchers that re-purpose off-the-shelf methods from the computer-science field of machine learning to create a "discovery engine" for data from randomized controlled trials (RCTs). The applied problem we seek to solve is that economists invest vast resources into carrying o…
New methods improve causal inference generalization using trial and observational data.
problem Limited trial data makes generalizing causal inferences to target populations statistically infeasible.
method Develops algorithms that combine trial and observational data to estimate complex nuisance functions.
result Improves generalization of causal inferences when the additional observational study is high-quality.
New findings show single-treatment effects are unidentifiable in factorial experiments.
problem Identifying the effect of a single intervention in factorial experiments.
method Formalized sufficient conditions for the identifiability of single-treatment effects and developed nonparametric sharp bounds.
result Researchers must justify assumptions for extrapolating single-treatment effects.
Deep neural networks (DNNs) are incredibly brittle due to adversarial examples. To robustify DNNs, adversarial training was proposed, which requires large-scale but well-labeled data. However, it is quite expensive to annotate large-scale data well. To compensate for this shortage, several seminal works are utilizing l…
Personalized medicine seeks to identify the causal effect of treatment for a particular patient as opposed to a clinical population at large. Most investigators estimate such personalized treatment effects by regressing the outcome of a randomized clinical trial (RCT) on patient covariates. The realized value of the ou…
Paper introduces EnCounteR for estimating causal effects using encouragement data.
problem Challenges in estimating causal effects due to incomplete randomization and limited encouragement data.
method Introduces a generalized IV estimator, EnCounteR, leveraging both observational and encouragement data.
result Demonstrates superior performance of EnCounteR over existing methods.
This study quantifies uncertainty in comparing treatments using RCTs with before-and-after measures.
problem Uncertainty in comparing treatments using RCTs with before-and-after measures.
method New statistical modeling principle called ETZ enables counterfactual uncertainty quantification (CUQ) in RCTs with Before-and-After Repeated Measures.
result CUQ typically has lower variability than factual uncertainty quantification and can be achieved in RCTs.
Paper introduces multi-scale methods to improve CATE estimation from EO data.
problem Challenges in balancing fine-grained and contextual information in EO-based causal inference.
method Multi-Scale Representation Concatenation, combining Vision Transformer and Causal Forests.
result Multi-scale approach captures effect heterogeneity better than single-scale models.
Novel approach to compute hazard ratios from observational studies using SCMs and backdoor adjustment.
problem Identifying causal relationships from observational data using hazard ratios.
method Backdoor adjustment through structural causal models (SCMs) and do-calculus.
result Novel approach for computing hazard ratios from observational studies.
Causal graph aids observational study insights in aSAH patients.
problem Lack of clear objectives and tools for identifying necessary adjustments in observational studies.
method Uses causal directed acyclic graphs (DAGs) to provide insights mid-study and identify necessary data enhancements.
result Midway insights and necessary data enhancements identified for meaningful causal questions.
Randomized Controlled Trials (RCTs) are the gold standard for comparing the effectiveness of a new treatment to the current one (the control). Most RCTs allocate the patients to the treatment group and the control group by uniform randomization. We show that this procedure can be highly sub-optimal (in terms of learnin…
We present PubMed 200k RCT, a new dataset based on PubMed for sequential sentence classification. The dataset consists of approximately 200,000 abstracts of randomized controlled trials, totaling 2.3 million sentences. Each sentence of each abstract is labeled with their role in the abstract using one of the following …
New method uses latent variables to estimate treatment effects from single-arm trials.
problem Estimating treatment effects from single-arm trials due to lack of external control groups.
method Latent-variable modeling with amortized variational inference for patient matching and direct effect estimation.
result Improved performance in direct treatment effect estimation and effect estimation via patient matching compared to previous methods.
Machine learning experiments often mislead due to unmet assumptions.
problem Machine learning experiments with pooled data may not meet necessary assumptions for unbiased causal effect estimation.
method Analysis of assumptions required for unbiased causal effect estimation in machine learning experiments.
result Practical applications of A/B-tests with machine learning models may not yield unbiased estimates of causal effect.
CausalSim corrects bias in trace-driven simulations for more accurate results.
problem Bias in trace-driven simulations due to system conditions during trace collection.
method CausalSim learns a causal model of system dynamics and latent factors from an RCT to remove bias from trace data.
result CausalSim reduces simulation errors by 53% and 61% compared to baselines, providing more accurate insights.
New method estimates optimal personalized treatment rules from mixed data sources.
problem Combining RCT and observational data for personalized treatment rules.
method Doubly robust estimator for value function, maximizing within pre-specified ITR class.
result Consistent and asymptotically normal optimal value estimator with N−1/3 rate of convergence. Study compares Cox model and RSF for predicting patient survival, finding RSF superior in certain scenarios.
problem Comparing predictive accuracy of Cox proportional hazards model and Random Survival Forest for patient-specific survival probabilities.
method Conducted a comprehensive comparison study using simulation scenarios and real-world datasets.
result RSF outperforms Cox model in nonproportional hazards settings and with treatment-covariate interactions.
New design detects confounders from treatment intent in ICU data.
problem Unobserved confounding in observational ICU data.
method Querying human experts to identify unobserved confounders.
result Demonstrates feasibility of detecting confounders from EHRs.
New approach estimates treatment effects from decentralized data.
problem Estimating treatment effects from multiple studies with limited data.
method Three classes of ATE estimators derived from Plug-in G-Formula.
result Asymptotic variance of estimators for linear models derived.
The paper proposes a method to find subgroups with significant treatment effects in noisy data.
problem Estimating the causal effects of interventions on noisy outcomes.
method A machine-learning method specifically optimized for finding subgroups with significant effects, designed to maximize the probability of obtaining a statistically significant positive treatment effect.
result The proposed method yields higher power in detecting subgroups affected by the treatment compared to standard tree-based tools.
CausalBench aims to advance causal learning research with a transparent platform.
problem Lack of unified benchmark datasets, algorithms, metrics, and evaluation interfaces for causal learning.
method Introduces CausalBench, a flexible benchmark framework for causal analysis and machine learning.
result Promotes scientific collaboration, reproducibility, and awareness in causal learning research.
The treatment effects of medications play a key role in guiding medical prescriptions. They are usually assessed with randomized controlled trials (RCTs), which are expensive. Recently, large-scale electronic health records (EHRs) have become available, opening up new opportunities for more cost-effective assessments. …
New method uses observational data to improve trial design efficiency.
problem Scarce randomized controlled trials; inefficiency of using observational data.
method Active Residual Learning, R-Design framework, R-EPIG criterion.
result Efficiently estimating residuals to correct observational bias improves trial design.
Q-Learner estimates ratio-based treatment effects without imposing parametric structures.
problem Estimating treatment effects as ratios in non-linear settings.
method Decomposes ratio-CATE into two classification tasks, using doubly robust augmentations.
result Q-Learner outperforms other methods in low-conversion and observational data settings.
New TTP framework fuses control arms while controlling Type-I error.
problem Bias in borrowing control data from previous trials.
method Kernel two-sample testing via MMD and equivalence testing.
result Higher power than standard TTP methods while maintaining error control.
New estimator improves ATT estimation efficiency with external controls.
problem Reduced efficiency when incorporating external controls into ATT estimation.
method Proposes a novel doubly robust estimator for ATT that maintains higher efficiency than standard approaches.
result Demonstrates improved efficiency of the new estimator compared to standard approaches, even under model misspecification.
G-computation improves clinical trial power with machine learning.
problem Balancing prognostic factors in randomized trials to prevent near-confounders.
method G-computation with penalized models (Lasso, Elasticnet) and algorithm-based methods (neural network, SVM, super learner).
result G-computation with Elasticnet and splines reduces variance and increases power in RCTs.
A new method flips class values to address class and treatment imbalance in uplift modeling and HTE.
problem Class and treatment imbalance in imbalanced RCT data.
method Class flipping approach to address imbalance without distorting predictions.
result The method does not distort predicted effects and does not require calibration.
New method minimizes decision errors in large treatment spaces.
problem Improving decision-making in large treatment spaces with biased observational data.
method Loss minimizes classification error of actions in large action space.
result Proves improved decision-making performance in large combinatorial action spaces.
Synthetic control method improves policy evaluation in high-dimensional settings.
problem Evaluating the impact of new policies in large-scale applications.
method Two-phase approach: nearest neighbor matching followed by supervised learning.
result The method successfully improves estimate accuracy in large-scale experiments.