arXiv research

Study improves statistical inference for CATEs using Lasso and DML.

DWTS uses observational data to improve clinical trial efficiency.

This paper introduces metrics for welfare analysis in dynamic models. We develop estimation and inference for these parameters even in the presence of a high-dimensional state space. Examples of welfare metrics include average welfare, average marginal welfare effects, and welfare decompositions into direct and indirec…

Javanmard and Montanari propose a debiased estimator for high-dimensional regression.

We simplify debiasing of LASSO estimates for practical computation.

Ridge regression is revisited with debiasing and thresholding, offering advantages over Lasso.

Proposes online debiasing to correct bias in adaptive data collection for high-dimensional linear regression.

Two approaches to directly estimating Riesz representer are shown to be numerically equivalent under certain conditions.

Paper proposes methods to reduce bias and variance in recommender systems.

Improved Lasso estimator speeds up variable selection.

Novel characterization of augmented balancing weights combining outcome and weighting models.

ADIGen: Automatic, Debiased, and Invariant Counterfactual Generation

Bayesian method corrects bias in treatment effect estimation.

The paper improves Lasso inference methods for survey data.

We devise a one-shot approach to distributed sparse regression in the high-dimensional setting. The key idea is to average "debiased" or "desparsified" lasso estimators. We show the approach converges at the same rate as the lasso as long as the dataset is not split across too many machines. We also extend the approach…

Contextual multi-armed bandit algorithms are widely used in sequential decision tasks such as news article recommendation systems, web page ad placement algorithms, and mobile health. Most of the existing algorithms have regret proportional to a polynomial function of the context dimension, $d$. In many applications ho…

The paper analyzes LASSO penalization for high-dimensional Beta regression models.

Corrects mismatch in consistency of nuisance estimators for doubly robust methods.

New algorithm for computing Wasserstein barycenters with guarantees.

Theory establishes optimal rates for estimating linear functionals without structural assumptions.

New method controls FDR for sparse GLMs, identifying positive and negative relationships.

We consider the problem of distributed multi-task learning, where each machine learns a separate, but related, task. Specifically, each machine learns a linear predictor in high-dimensional space,where all tasks share the same small support. We present a communication-efficient estimator based on the debiased lasso and…

We study sparse group Lasso for high-dimensional double sparse linear regression, where the parameter of interest is simultaneously element-wise and group-wise sparse. This problem is an important instance of the simultaneously structured model -- an actively studied topic in statistics and machine learning. In the noi…

In this study, we propose an automatic learning method for variables selection based on Lasso in epidemiology context. One of the aim of this approach is to overcome the pretreatment of experts in medicine and epidemiology on collected data. These pretreatment consist in recoding some variables and to choose some inter…

This paper investigates robust and efficient DR/RDR estimators for WATEs.

Functional brain networks are well described and estimated from data with Gaussian Graphical Models (GGMs), e.g. using sparse inverse covariance estimators. Comparing functional connectivity of subjects in two populations calls for comparing these estimated GGMs. Our goal is to identify differences in GGMs known to hav…

Computing partition functions, the normalizing constants of probability distributions, is often hard. Variants of importance sampling give unbiased estimates of a normalizer Z, however, unbiased estimates of the reciprocal 1/Z are harder to obtain. Unbiased estimates of 1/Z allow Markov chain Monte Carlo sampling of "d…

Study examines inference methods after variable selection in Cox models.

Develops a new method for uncertainty quantification in high-dimensional learning.

New method for estimating treatment effects without complex propensity models.

Study high-dimensional Granger causality tests for VIX and financial news.

This paper introduces a novel online inference method for high-dimensional GLMs.

ADML combines debiased learning with data-driven model selection for efficient inference.

Develops a method to estimate average hazard under non-proportional hazards without relying on proportional hazards assumption.

DoubleGen addresses bias in generative modeling of counterfactuals.

DebiNet uses over-parameterized neural networks to improve linear model performance and debiasing.

The Lasso method is analyzed for high-dimensional regression with Gaussian designs, leading to new insights on its performance.

We consider linear regression in the high-dimensional regime where the number of observations $n$ is smaller than the number of parameters $p$. A very successful approach in this setting uses $\ell_1$-penalized least squares (a.k.a. the Lasso) to search for a subset of $s_0< n$ parameters that best explain the data, wh…

A new method for high-dimensional $M$-estimation with missing outcomes.

A new method for feature selection robust to noise and design variability.

Estimates causal effects using machine learning for binary treatment and mediator.

Proposes RCL method to improve ATE estimation from observational data.

Improves robustness of propensity score estimators in challenging settings.

Semiparametric method removes bias in functional bilevel gradient estimation.

Performing statistical inference in high-dimension is an outstanding challenge. A major source of difficulty is the absence of precise information on the distribution of high-dimensional estimators. Here, we consider linear regression in the high-dimensional regime $p\gg n$. In this context, we would like to perform in…

Paper introduces structured sparsity estimators for Generalized Linear Models.

High-dimensional inference for sparse spectral precision matrices

Method estimates dynamic treatment effects using machine learning and g-estimation.