The paper introduces a plug‑in estimator for the non‑iterative conditional expectation (NICE) g‑formula that uses long short‑term memory (LSTM) models instead of traditional parametric models. It demonstrates, through a complex data‑generation scenario resembling an HIV cohort, that the LSTM‑based estimator yields lower bias and near‑nominal 95% confidence‑interval coverage compared to parametric alternatives. The authors suggest this work as a step toward a theoretical framework for deep‑learning plug‑in estimators in high‑dimensional longitudinal causal inference.
arXiv:2606. 05797v1 Announce Type: new Abstract: Longitudinal treatment decisions require predicting potential outcomes under future treatment sequences in the presence of time-varying confounding, heterogeneous patient dynamics, and limited domain-specific data.
The paper introduces new one‑step and targeted minimum loss‑based estimators for the average treatment effect and the average treatment effect on the treated under the front‑door criterion. These estimators use multiple parameterizations of the observed data distribution, can avoid modeling the mediator density, and are compatible with flexible machine‑learning nuisance estimation. The authors derive conditions for root‑n consistency, asymptotic linearity, and develop doubly robust tests for identification assumptions, demonstrating improved efficiency and favorable performance in simulations and real‑world applications in education and emergency medicine.
The paper introduces PICLE-PFN, a foundation model for longitudinal causal effect estimation that handles dynamic treatment policies. It represents policies via in-context history-decision demonstrations and is pretrained on synthetic data from longitudinal structural causal models, enabling it to predict conditional average potential outcomes in a single forward pass without task-specific training. Experiments on synthetic and semi-synthetic benchmarks show competitive RMSE and better uncertainty calibration while reducing deployment runtime compared to conventional estimators.
arXiv:2609.06294v1 Announce Type: new Abstract: Estimating conditional average treatment effects (CATE) enables efficient targeting of interventions, but many applications have limited experimental s...
arXiv:2607. 14940v1 Announce Type: new Abstract: We study causal inference under outcome interference for sequential, observational settings.
The paper tackles the problem of estimating causal effects when an unobserved confounder is present. It assumes a single, possibly multi‑dimensional proxy variable for the confounder and knowledge of the mechanism that generates this proxy. Under the Single Proxy Identifiability of Causal Effects (SPICE) assumption, the authors prove that the error mechanism is complete and causal effects are identifiable, extending prior proxy‑based results to continuous, multi‑dimensional settings and more flexible functional forms. They also introduce SPICE‑Net, a neural‑network‑based framework for estimating causal effects applicable to both discrete and continuous treatments.
The paper proposes supervised learning objectives for estimating causal effects of high‑dimensional treatments, such as text or continuous variables, by recasting causal inference as a learning problem under realizability and no unobserved confounding. It shows that causal error can be decomposed into moment‑balancing errors of increasing order and designs objectives that directly improve causal estimation by exploiting the structure of conditional outcomes. The authors also demonstrate how to project high‑dimensional treatment effects onto coarser attributes, enabling a single model to answer multiple causal questions without extra attribute‑specific training, and validate their approach empirically across continuous, discrete, and text treatments.
arXiv:2611.08077v1 Announce Type: cross Abstract: We propose a doubly-robust method to estimate the causal effects of multi-valued treatment regimes on patient outcomes. A multi-valued treatment regi...
arXiv:2606. 01539v1 Announce Type: cross Abstract: Estimating the causal effect of time-varying treatments on survival outcomes in large observational studies is computationally demanding, particularly when outcomes are rare.
arXiv:2604. 16763v3 Announce Type: replace Abstract: Causal inference from electronic health records (EHR) is fundamentally limited by unmeasured confounding: critical clinical states such as frailty, goals of care, and mental status are documented in free-text notes but absent from structured data.
The paper introduces a method for estimating the causal effects of T cell receptor (TCR) sequences on patient outcomes using observational TCR sequencing and clinical data. It corrects for unobserved confounders by leveraging the pre-selection TCR repertoire generated through V(D)J recombination as a natural experiment, and employs permutation‑invariant neural networks to scale to millions of sequences. The approach is validated on semisynthetic data and applied to COVID‑19 severity, identifying TCRs that are observed in patients, bind SARS‑CoV‑2 antigens in vitro, and positively influence clinical outcomes.
arXiv:2608. 10339v1 Announce Type: cross Abstract: Hospital quality improvement (QI) programs routinely face multiple candidate interventions to optimize hospital flow, but existing methods struggle to estimate and rank the causal effects of such interventions.