arXiv research

New method tackles confounded bandit problems with dual instrumental variables.

Study linear contextual bandits with confounded offline data, improving regret bounds.

A new algorithm CAP learns optimal policies from observational data with confounding bias and missing observations.

DFPV improves PCL for confounded bandit policy evaluation.

Bayesian optimization improved for biased data.

The paper tackles bandit problems with biased offline data by using causal methods.

Motivated by online recommendation and advertising systems, we consider a causal model for stochastic contextual bandits with a latent low-dimensional confounder. In our model, there are $L$ observed contexts and $K$ arms of the bandit. The observed context influences the reward obtained through a latent confounder var…

New algorithm learns causal graph to minimize regret in bandits without full structure.

Sharp policy value estimation for contextual bandits with unobserved confounders.

A new framework estimates policy value robustly against confounders.

This paper studies semiparametric contextual bandits, a generalization of the linear stochastic bandit problem where the reward for an action is modeled as a linear function of known action features confounded by an non-linear action-independent term. We design new algorithms that achieve $\tilde{O}(d\sqrt{T})$ regret …

Evaluating novel contextual bandit policies using logged data is crucial in applications where exploration is costly, such as medicine. But it usually relies on the assumption of no unobserved confounders, which is bound to fail in practice. We study the question of policy evaluation when we instead have proxies for th…

Bayesian bandits misspecification affects UX optimization, revealing new models.

New algorithms exploit mean bounds to improve bandit problem performance.

New algorithm identifies best intervention without graph knowledge.

Applying causal inference models in areas such as economics, healthcare and marketing receives great interest from the machine learning community. In particular, estimating the individual-treatment-effect (ITE) in settings such as precision medicine and targeted advertising has peaked in application. Optimising this IT…

The paper proposes a method to precisely decompose confounders and estimate treatment effects.

Research shows continual learning challenges in confounded datasets.

Statistical learning on biological data can be challenging due to confounding variables in sample collection and processing. Confounders can cause models to generalize poorly and result in inaccurate prediction performance metrics if models are not validated thoroughly. In this paper, we propose methods to control for …

Causal inference from observational data requires assumptions. These assumptions range from measuring confounders to identifying instruments. Traditionally, causal inference assumptions have focused on estimation of effects for a single treatment. In this work, we construct techniques for estimation with multiple treat…

Detect hidden confounding in observational data using multiple environments.

Machine learning confound removal biases results, leading to misleading predictions.

KRCD detects unobserved confounders in nonlinear observational data.

Study identifies conditions for proxy adjustment in confounded binary treatment outcomes.

New method estimates policy performance under unobserved confounding.

Consistent estimator derived for confounding strength in observational data.

A new method uses randomized trials to estimate the strength of unobserved confounding.

This paper tackles confounding biases in data augmentation.

New method recovers latent confounders from high-dimensional proxy variables.

Proposes tests to control confounding bias in predictive models.

New approach for causal inference with interdependent, time-varying latent confounders.

Causal discovery from data affected by latent confounders is an important and difficult challenge. Causal functional model-based approaches have not been used to present variables whose relationships are affected by latent confounders, while some constraint-based methods can present them. This paper proposes a causal f…

Estimating the individual treatment effect (ITE) from observational data is essential in medicine. A central challenge in estimating the ITE is handling confounders, which are factors that affect both an intervention and its outcome. Most previous work relies on the unconfoundedness assumption, which posits that all th…

CARE improves LLM aggregation by accounting for shared confounders.

Proposes BDCM to handle unmeasured confounders in causal inference.

New methods prioritize acquiring confounding features for efficient treatment effect estimation.

Valid causal inference with unobserved confounding in high-dimensional settings.

Spectral deconfounding improves machine learning models by reducing hidden confounding effects.

We propose a method for inferring the existence of a latent common cause ('confounder') of two observed random variables. The method assumes that the two effects of the confounder are (possibly nonlinear) functions of the confounder plus independent, additive noise. We discuss under which conditions the model is identi…

Study shows how adjusting for a binary proxy can bound causal effects.

Paper tackles unobserved confounding in human-AI collaborations.

SpaCE tackles spatial confounding in scientific studies.

Estimates causal effects with selection bias and confounding using regression.

Paper tackles backwards-compatible data adaptation for confounded covariate and label shifts.

Deep CITs test conditional independence in images, improving brain MRI scan analysis.

In this paper, we study the confounder detection problem in the linear model, where the target variable $Y$ is predicted using its $n$ potential causes $X_n=(x_1,...,x_n)^T$. Based on an assumption of rotation invariant generating process of the model, recent study shows that the spectral measure induced by the regress…

Proposes DSW for unbiased ITE estimation with dynamic confounders.

Paper uses Gaussian processes to handle shared latent confounders in causal inference.