arXiv research

The paper argues for using Neyman orthogonal score for balancing in debiased machine learning.

New method balances covariates for stable causal survival effect estimation.

Unified theory for causal inference using various methods.

Dynamic treatment effects estimated over time using covariate balancing.

The paper proposes a new method for covariate balancing using IPM to improve causal inference.

A new algorithm speeds up rerandomization for better experiment balance.

We study optimal covariate balance for causal inferences from observational data when rich covariates and complex relationships necessitate flexible modeling with neural networks. Standard approaches such as propensity weighting and matching/balancing fail in such settings due to miscalibrated propensity nets and inapp…

Cross-balancing improves causal inference by balancing features with outcome data.

Estimates causal effects using neural networks for balancing covariates.

A new method corrects bias in causal inference by balancing covariate distributions.

SVM used for estimating treatment effects without confounding.

We determine the Christoffel's symbols for the Siegel-Jacobi ball endowed with the balanced metric. We study the equations of geodesics on the Siegel-Jacobi ball. We calculate the covariant derivative of one-forms in the variables in which is expressed the balanced metric on the Siegel-Jacobi ball.

Differentially private method for estimating individualized treatment rules.

The purpose of this paper is to extend the Green-Naghdi-Rivlin balance of energy method to continua with microstructure. The key idea is to replace the group of Galilean transformations with the group of diffeomorphisms of the ambient space. A key advantage is that one obtains in a natural way all the needed balance la…

Paper tackles CATE estimation with missing treatment info.

New method for causal inference with complex treatment compositions.

Improves representation learning for individual treatment effect estimation.

Study proposes a new method to estimate bias-correction term for ATE estimation.

We introduce a new normalization technique that exhibits the fast convergence properties of batch normalization using a transformation of layer weights instead of layer outputs. The proposed technique keeps the contribution of positive and negative weights to the layer output balanced. We validate our method on a set o…

Learning causal effects from observational data greatly benefits a variety of domains such as health care, education and sociology. For instance, one could estimate the impact of a new drug on specific individuals to assist the clinic plan and improve the survival rate. In this paper, we focus on studying the problem o…

Develops a direct debiased machine learning framework using Bregman divergence.

New method for estimating covariance with robustness to outliers.

TV-SurvCaus improves causal inference for dynamic treatments in survival analysis.

Kernel balancing weights are generalized as KRRR, providing better confidence intervals for treatment effects.

Many scientific questions require estimating the effects of continuous treatments. Outcome modeling and weighted regression based on the generalized propensity score are the most commonly used methods to evaluate continuous effects. However, these techniques may be sensitive to model misspecification, extreme weights o…

Unified framework for debiased machine learning using Riesz representer and Bregman divergence.

We analyze geometrical structures necessary to represent bulk and surface interactions of standard and substructural nature in complex bodies. Our attention is mainly focused on the influence of diffuse interfaces on sharp discontinuity surfaces. In analyzing this phenomenon, we prove the covariance of surface balances…

The paper proposes a method to estimate treatment effects using CAR designs with additional covariates.

SPARKLE handles high-dimensional covariates for online decision-making.

New method learns decisions from collective preferences without individual covariates.

sgboost reduces variable selection bias in boosting with balanced group selection.

A new IPM uses ReLU networks to measure probability discrepancies.

Covariance-Driven Regression Trees reduce overfitting in CART.

Proposes a method to improve learning when training data is not representative.

DiPriMe forests use private medians to create balanced tree splits for privacy-protected data.

In this paper, we propose a robust method to estimate the average treatment effects in observational studies when the number of potential confounders is possibly much greater than the sample size. We first use a class of penalized M-estimators for the propensity score and outcome models. We then calibrate the initial e…

Efficiently solves large portfolio optimization problems by reducing and sparsifying covariance matrices.

In the absence of unobserved confounders, matching and weighting methods are widely used to estimate causal quantities including the Average Treatment Effect on the Treated (ATT). Unfortunately, these methods do not necessarily achieve their goal of making the multivariate distribution of covariates for the control gro…

Classical mean-variance portfolio theory tells us how to construct a portfolio of assets which has the greatest expected return for a given level of return volatility. Utility theory then allows an investor to choose the point along this efficient frontier which optimally balances her desire for excess expected return …

A new model integrates covariates with grade of membership analysis for better latent structure recovery.

In this contribution we present an intrinsic description of time-variant Port Hamiltonian systems as they appear in modeling and control theory. This formulation is based on the splitting of the state bundle and the use of appropriate covariant derivatives, which guarantees that the structure of the equations is invari…

FSRL balances fairness and sufficiency in learning representations.

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

We consider the problem of how to assign treatment in a randomized experiment, in which the correlation among the outcomes is informed by a network available pre-intervention. Working within the potential outcome causal framework, we develop a class of models that posit such a correlation structure among the outcomes. …

New scalable method balances hospital profit status and heart attack outcomes.

BICauseTree improves causal effect estimation by identifying clusters and balancing treatment allocation.

Hybrid ResNet and RMT improve covariance matrix estimation for cryptocurrency portfolios.

CInA method uses attention to improve causal inference.