1期

最新文章

• Boosting the power of kernel two-sample tests AI翻译
  作者：Chatterjee, A.; Bhattacharya, B. B.
  作者单位：University of Pennsylvania
  摘要：The kernel two-sample test based on the maximum mean discrepancy is one of the most popular methods for detecting differences between two distributions over general metric spaces. In this paper we propose a method to boost the power of the kernel test by combining maximum mean discrepancy estimates over multiple kernels using their Mahalanobis distance. We derive the asymptotic null distribution of the proposed test statistic and use a multiplier bootstrap approach to efficiently compute the r...
• Rejoinder: Causal inference with misspecified exposure mappings: separating definitions and assumptions AI翻译
  作者：Saevje, F.
  作者单位：Yale University
• Robust sample weighting to facilitate individualized treatment rule learning for a target population AI翻译
  作者：Chen, Rui; Huling, Jared D.; Chen, Guanhua; Yu, Menggang
  作者单位：University of Wisconsin System; University of Wisconsin Madison; University of Minnesota System; University of Minnesota Twin Cities; University of Wisconsin System; University of Wisconsin Madison
  摘要：Learning individualized treatment rules is an important topic in precision medicine. Current literature mainly focuses on deriving individualized treatment rules from a single source population. We consider the observational data setting when the source population differs from a target population of interest. Compared with causal generalization for the average treatment effect that is a scalar quantity, individualized treatment rule generalization poses new challenges due to the need to model ...
• Universal robust regression via maximum mean discrepancy AI翻译
  作者：Alquier, P.; Gerber, M.
  作者单位：ESSEC Business School; University of Bristol
  摘要：Many modern datasets are collected automatically and are thus easily contaminated by outliers. This has led to a renewed interest in robust estimation, including new notions of robustness such as robustness to adversarial contamination of the data. However, most robust estimation methods are designed for a specific model. Notably, many methods were proposed recently to obtain robust estimators in linear models, or generalized linear models, and a few were developed for very specific settings, ...
• With random regressors, least squares inference is robust to correlated errors with unknown correlation structure AI翻译
  作者：Zhang, Zifeng; Ding, Peng; Zhou, Wen; Wang, Haonan
  作者单位：Colorado State University System; Colorado State University Fort Collins; University of California System; University of California Berkeley; New York University
  摘要：Linear regression is arguably the most widely used statistical method. With fixed regressors and correlated errors, the conventional wisdom is to modify the variance-covariance estimator to accommodate the known correlation structure of the errors. We depart from existing literature by showing that with random regressors, linear regression inference is robust to correlated errors with unknown correlation structure. The existing theoretical analyses for linear regression are no longer valid bec...
• Estimating causal effects under non-individualistic treatments due to network entanglement AI翻译
  作者：Toulis, P.; Volfovsky, A.; Airoldi, E. M.
  作者单位：University of Chicago; Duke University; Pennsylvania Commonwealth System of Higher Education (PCSHE); Temple University
  摘要：In many observational studies, the treatment assignment mechanism is not individualistic, as it allows the probability of treatment of a unit to depend on quantities beyond the unit's covariates. In such settings, unit treatments may be entangled in complex ways. In this article, we consider a particular instance of this problem where the treatments are entangled by a social network among units. For instance, when studying the effects of peer interaction on a social media platform, the treatme...
• An adaptive null proportion estimator for false discovery rate control AI翻译
  作者：Gao, Zijun
  作者单位：University of Southern California
  摘要：The false discovery rate is a commonly used criterion in multiple testing, and the Benjamini-Hochberg procedure is a standard approach to false discovery rate control. To increase its power, adaptive Benjamini-Hochberg procedures, that use estimates of the null proportion, have been proposed. A particularly popular approach being that based on Storey's estimator. The performance of Storey's estimator hinges on a critical hyperparameter, such that a pre-fixed configuration may lack power and ex...
• Bayesian learning of network structures from interventional experimental data AI翻译
  作者：Castelletti, F.; Peluso, S.
  作者单位：Catholic University of the Sacred Heart; University of Milano-Bicocca
  摘要：Directed acyclic graphs provide an effective framework for learning causal relationships among variables given multivariate observations. Under pure observational data, directed acyclic graphs encoding the same conditional independencies cannot be distinguished and are collected into Markov equivalence classes. In many contexts, however, observational measurements are supplemented by interventional data that improve directed acyclic graph identifiability and enhance causal effect estimation. W...
• Efficient evaluation of natural stochastic policies in off-line reinforcement learning AI翻译
  作者：Kallus, Nathan; Uehara, Masatoshi
  作者单位：Cornell University; Cornell University
  摘要：We study the efficient off-policy evaluation of natural stochastic policies, which are defined in terms of deviations from the unknown behaviour policy. This is a departure from the literature on off-policy evaluation that largely considers the evaluation of explicitly specified policies. Crucially, off-line reinforcement learning with natural stochastic policies can help alleviate issues of weak overlap, lead to policies that build upon current practice and improve policies' implementability ...
• Scalable subsampling: computation, aggregation and inference AI翻译
  作者：Politis, Dimitris
  作者单位：University of California System; University of California San Diego
  摘要：Subsampling has seen a resurgence in the big data era where the standard, full-resample size bootstrap can be infeasible to compute. Nevertheless, even choosing a single random subsample of size b can be computationally challenging with both b and the sample size n being very large. This paper shows how a set of appropriately chosen, nonrandom subsamples can be used to conduct effective, and computationally feasible, subsampling distribution estimation. Furthermore, the same set of subsamples ...