doubly robust
Recently Published Documents


TOTAL DOCUMENTS

250
(FIVE YEARS 100)

H-INDEX

22
(FIVE YEARS 3)

2022 ◽  
Vol 73 ◽  
pp. 209-229
Author(s):  
Chong Liu ◽  
Yu-Xiang Wang

Large-scale labeled dataset is the indispensable fuel that ignites the AI revolution as we see today. Most such datasets are constructed using crowdsourcing services such as Amazon Mechanical Turk which provides noisy labels from non-experts at a fair price. The sheer size of such datasets mandates that it is only feasible to collect a few labels per data point. We formulate the problem of test-time label aggregation as a statistical estimation problem of inferring the expected voting score. By imitating workers with supervised learners and using them in a doubly robust estimation framework, we prove that the variance of estimation can be substantially reduced, even if the learner is a poor approximation. Synthetic and real-world experiments show that by combining the doubly robust approach with adaptive worker/item selection rules, we often need much lower label cost to achieve nearly the same accuracy as in the ideal world where all workers label all data points.


2022 ◽  
Vol 0 (0) ◽  
Author(s):  
Asma Bahamyirou ◽  
Mireille E. Schnitzer ◽  
Edward H. Kennedy ◽  
Lucie Blais ◽  
Yi Yang

Abstract Effect modification occurs when the effect of a treatment on an outcome differsaccording to the level of some pre-treatment variable (the effect modifier). Assessing an effect modifier is not a straight-forward task even for a subject matter expert. In this paper, we propose a two-stageprocedure to automatically selecteffect modifying variables in a Marginal Structural Model (MSM) with a single time point exposure based on the two nuisance quantities (the conditionaloutcome expectation and propensity score). We highlight the performance of our proposal in a simulation study. Finally, to illustrate tractability of our proposed methods, we apply them to analyze a set of pregnancy data. We estimate the conditional expected difference in the counterfactual birth weight if all women were exposed to inhaled corticosteroids during pregnancy versus the counterfactual birthweight if all women were not, using data from asthma medications during pregnancy.


2021 ◽  
Author(s):  
SHANTANU GHOSH ◽  
Zheng Feng ◽  
Jiang Bian ◽  
Kevin Butler ◽  
Mattia Prosperi

Abstract Determining causal effects of interventions onto outcomes from observational (non-randomized) data, e.g., treatment repurposing using electronic health records, is challenging due to underlying bias. Causal deep learning has improved over traditional techniques for estimating individualized treatment effects. We present the Doubly Robust Variational Information-theoretic Deep Adversarial Learning (DR-VIDAL), a novel generative framework that combines two joint models of treatment and outcome, ensuring an unbiased estimation even when one of the two is misspecified. DR-VIDAL uses a variational autoencoder (VAE) to factorize confounders into latent variables according to causal assumptions; then, an information-theoretic generative adversarial network (Info-GAN) is used to generate counterfactuals; finally, a doubly robust block incorporates propensity matching/weighting into predictions. On synthetic and real-world datasets, DR-VIDAL achieves better performance than other non-generative and generative methods. In conclusion, DR-VIDAL uniquely fuses causal assumptions, VAE, Info-GAN, and doubly robustness into a comprehensive, performant framework. Code is available at: https://bitbucket.org/goingdeep2406/dr-vidal/src/master/


2021 ◽  
pp. 096228022110473
Author(s):  
Arthur Chatton ◽  
Florent Le Borgne ◽  
Clémence Leyrat ◽  
Yohann Foucher

In time-to-event settings, g-computation and doubly robust estimators are based on discrete-time data. However, many biological processes are evolving continuously over time. In this paper, we extend the g-computation and the doubly robust standardisation procedures to a continuous-time context. We compare their performance to the well-known inverse-probability-weighting estimator for the estimation of the hazard ratio and restricted mean survival times difference, using a simulation study. Under a correct model specification, all methods are unbiased, but g-computation and the doubly robust standardisation are more efficient than inverse-probability-weighting. We also analyse two real-world datasets to illustrate the practical implementation of these approaches. We have updated the R package RISCA to facilitate the use of these methods and their dissemination.


2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Luming Zhang ◽  
Fengshuo Xu ◽  
Didi Han ◽  
Tao Huang ◽  
Shaojin Li ◽  
...  

Abstract Background Sepsis-associated acute kidney injury (S-AKI) is a common and life-threatening complication in hospitalized and critically ill patients. This condition is an independent cause of death. This study was performed to investigate the correlation between the trajectory of urine output within 24 h and S-AKI. Methods Patients with sepsis were studied retrospectively based on the Medical Information Mart for Intensive Care IV. Latent growth mixture modeling was used to classify the trajectory of urine output changes within 24 h of sepsis diagnosis. The outcome of this study is AKI that occurs 24 h after sepsis. Cox proportional hazard model, Fine–Gray subdistribution proportional hazard model, and doubly robust estimation method were used to explore the risk of AKI in patients with different trajectory classes. Results A total of 9869 sepsis patients were included in this study, and their 24-h urine output trajectories were divided into five classes. The Cox proportional hazard model showed that compared with class 1, the HR (95% CI) values for classes 3, 4, and 5 were 1.460 (1.137–1.875), 1.532 (1.197–1.961), and 2.232 (1.795–2.774), respectively. Competing risk model and doubly robust estimation methods reached similar results. Conclusions The trajectory of urine output within 24 h of sepsis patients has a certain impact on the occurrence of AKI. Therefore, in the early treatment of sepsis, close attention should be paid to changes in the patient's urine output to prevent the occurrence of S-AKI.


2021 ◽  
Author(s):  
Nicole M. Butera ◽  
Donglin Zeng ◽  
Annie Green Howard ◽  
Penny Gordon‐Larsen ◽  
Jianwen Cai

2021 ◽  
pp. 107755872110527
Author(s):  
Samuel H. Masters ◽  
Regina I. Rutledge ◽  
Marisa Morrison ◽  
Heather A. Beil ◽  
Susan G. Haber

There is little evidence regarding population equity in alternative payment models (APMs). We aimed to determine whether one such APM, the Maryland All-Payer Model (MDAPM), had differential effects on subpopulations of vulnerable Medicare beneficiaries. We utilized Medicare fee-for-service claims for beneficiaries living in Maryland and 48 comparison hospital market areas between 2011 and 2018. We used doubly robust difference-in-difference-in-differences regression models to estimate the differential effects of MDAPM on Medicare beneficiaries by dual eligibility for Medicare and Medicaid, disability as original reason for Medicare entitlement, presence of multiple chronic conditions (MCC), race, and rural residency status. Dual, disabled, and beneficiaries with MCC had greater reductions in expenditures and utilization than their counterparts. Hospitals may have prioritized high-cost, high-need patients as they changed their care delivery practices. The percentage of hospital discharges with 14-day follow-up was significantly lower for disadvantaged subpopulations, including duals, disabled, and non-White.


Sign in / Sign up

Export Citation Format

Share Document