研究园简易版 · 经典书籍 · 期刊文章 · 有原文 · 完整版(可多选筛选、搜索) · 首页
2026 年的期刊文章 · 第 42 页
共 1756 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2026 到 2026。
按期刊、按主题看 →
SiM Causal Inference Meta-Analysis & Data Integration Longitudinal Data · 发表于 2026,收录于 2026-08-17
Nan Qiao, Jingxiao Zhang, Zishu Zhan
摘要
Estimation of optimal individualized treatment rules (ITRs) tailored to patient characteristics is a crucial component of precision medicine. When such problems arise in multicenter randomized trials with known treatment assignment probabilities, the optimal ITR may vary across sites because of site-level heterogeneity, and directly pooling data across centers can lead to biased ITR estimation. Meanwhile, privacy constraints often prevent investigators from accessing individual-level data from all centers. We therefore propose Heterogeneous Distributed Learning (HD-learning) to estimate optimal site-specific ITRs using only summary-level data from each site in a single communication round. The method uses a distributed mixed-effects model to accommodate common fixed effects and site-specific random effects. We further develop Shrinkage HD-learning (SHD-learning) for moderate- or high-dimensional covariates. Under suitable conditions, we establish asymptotic properties of the proposed estimators. Simulations show that our methods are lossless relative to the corresponding pooled estimator and outperform methods that ignore heterogeneity. We illustrate the proposed methods using the acute upper respiratory tract infections (AURTIs) study and the opioid withdrawal study.
JMLR Other · 发表于 2026
Siming Zheng, Guohao Shen, Yuanyuan Lin, Jian Huang
摘要
We consider the problem of density-ratio estimation using Bregman Divergence with Deep ReLU feedforward neural networks (BDD). We establish non-asymptotic error bounds for BDD density-ratio estimators, which are minimax optimal up to a logarithmic factor when the data distribution has finite support. As an application of our theoretical findings, we propose an estimator for the KL-divergence that is asymptotically normal, leveraging our convergence results for the deep density-ratio estimator and a data-splitting method. We also extend our results to cases with unbounded support and unbounded density ratios. Furthermore, we show that the BDD density-ratio estimator can mitigate the curse of dimensionality when data distributions are supported on an approximately low-dimensional manifold. Our results are applied to investigate the convergence properties of the telescoping density-ratio estimator proposed by Rhodes (2020). We provide sufficient conditions under which it achieves a lower error bound than a single-ratio estimator. Moreover, we conduct simulation studies to validate our main theoretical results and assess the performance of the BDD density-ratio estimator.
JMLR Machine Learning · 发表于 2026
Jing Wang, Fengzhuo Zhang, Xiaoli Li, Vincent Y.~ F. Tan, Tianyu Pang, Chao Du, Aixin Sun, Zhuoran Yang
摘要
Auto-Regressive Video Diffusion Models (AR-VDMs) have shown strong capabilities in generating long, photorealistic videos, but suffer from two key limitations: (i) history forgetting, where the model loses track of previously generated content, and (ii) temporal degradation, where frame quality deteriorates over time. Yet a rigorous theoretical analysis of these phenomena is lacking, and existing empirical understanding remains insufficiently grounded. In this paper, we introduce Meta-ARVDM, a unified analytical framework that studies both errors through the shared autoregressive structure of AR-VDMs. We show that history forgetting is characterized by the conditional mutual information between the generated output and preceding frames, conditioned on inputs, and prove that incorporating more past frames monotonically alleviates history forgetting, thereby theoretically justifying a common belief in existing works. Moreover, our theory reveals that standard metrics fail to capture this effect, motivating a new evaluation protocol based on a “needle-in-a-haystack” task in closed-ended environments (DMLab and Minecraft). We further show that temporal degradation can be quantified by the cumulative sum of per-step errors, enabling prediction of degradation for different schedulers without video rollout. Finally, our evaluation uncovers a strong empirical correlation between history forgetting and temporal degradation, a connection not previously reported.
Econometrica Econometrics & Economics · 发表于 2026
Wojciech Olszewski, Philip J. Reny, Ron Siegel
摘要
We provide sufficient conditions for equilibrium existence in first‐price auctions with private values that accommodate non quasi‐linear utilities and value‐distributions that contain atoms and exhibit positive or negative correlation. These conditions show that equilibrium existence often turns on properties of a single statistic of the joint distribution of values, namely, the minimum value in the support of the high‐value distribution (the mHV). We also show that modifying the standard tie‐breaking rule only at the mHV is enough to guarantee equilibrium existence without our sufficient conditions. Our results also apply to Bertrand price competition when each firm's constant marginal cost is private information.
JMLR Machine Learning · 发表于 2026
Xinyu Tian, Xiaotong Shen
摘要
This paper investigates the accuracy of generative models and the impact of knowledge transfer on their generation precision. Specifically, we examine a generative model for a target task, fine-tuned using a pre-trained model from a source task. Building on the "Shared Embedding" concept, which bridges the source and target tasks, we introduce a novel framework for transfer learning under distribution metrics such as the Kullback-Leibler divergence. This framework underscores the importance of leveraging inherent similarities between diverse tasks despite their distinct data distributions. Our theory suggests that the shared structures can augment the generation accuracy for a target task, reliant on the capability of a source model to identify shared structures and effective knowledge transfer from source to target learning. To demonstrate the practical utility of this framework, we explore the theoretical implications for two specific generative models: diffusion and normalizing flows. The results show enhanced performance in both models over their non-transfer counterparts, indicating advancements for diffusion models and providing fresh insights into normalizing flows in transfer and non-transfer settings. These results highlight the significant contribution of knowledge transfer in boosting the generation capabilities of these models.
JMLR Machine Learning Networks & Graphs Testing & Uncertainty · 发表于 2026
Yidong Zhou, Su I Iao, Hans-Georg Müller
摘要
Many modern applications involve predicting structured, non-Euclidean outputs such as probability distributions, networks, and symmetric positive-definite matrices. These outputs are naturally modeled as elements of general metric spaces, where classical regression techniques that rely on vector space structure no longer apply. We introduce E2M (End-to-End Metric regression), a deep learning framework for predicting metric space-valued outputs. E2M performs prediction via weighted Fréchet means over training outputs, where the weights are learned by a neural network conditioned on the input. This construction provides a principled mechanism for geometry-aware prediction that avoids surrogate embeddings and restrictive parametric assumptions, while fully preserving the intrinsic geometry of the output space. We establish theoretical guarantees, including a universal approximation theorem that characterizes the expressive capacity of the model and a convergence analysis of the entropy-regularized training objective. Through extensive simulations involving probability distributions, networks, and symmetric positive-definite matrices, we show that E2M consistently achieves state-of-the-art performance, with its advantages becoming more pronounced at larger sample sizes. Applications to human mortality distributions and New York City taxi networks further demonstrate the flexibility and practical utility of this framework.
Econometrica Bayesian Methods · 发表于 2026
Jiafeng Chen
摘要
Gaussian empirical Bayes methods usually maintain a precision independence assumption: The unknown parameters of interest are independent from the known standard errors of the estimates. This assumption is often theoretically questionable and empirically rejected. This paper proposes to model the conditional distribution of the parameter given the standard errors as a flexibly parameterized location‐scale family of distributions, leading to a family of methods that we call close . The close framework unifies and generalizes several proposals under precision dependence. We argue that the most flexible member of the close family is a minimalist and computationally efficient default for accounting for precision dependence. We analyze this method and show that it is competitive in terms of the regret of subsequent decision rules. Empirically, using close leads to sizable gains for selecting high‐mobility Census tracts.
JMLR Spatial & Time Series Networks & Graphs Causal Inference · 发表于 2026
Jae Ho Chang, Subhadeep Paul
摘要
We propose an Embedding Network Autoregressive Model for multivariate networked longitudinal data. We assume the network is generated from a latent variable model, and these unobserved variables are included in a structural peer effect model or a time series network autoregressive model. This approach takes a unified view of two related yet different problems: (1) modeling and predicting multivariate networked time series data and (2) causal peer influence estimation in the presence of confounding due to homophily from finite-time longitudinal data. Our estimation strategy comprises estimating latent variables from the observed network, followed by least squares estimation of the network autoregressive model. We show that the momentum and peer effect parameters estimated with our method are consistent and asymptotically normally distributed in setups with a growing number of network vertices ($N$) while considering both a growing number of time points $T$ (for the time series problem) and finite $T$ cases (for the peer effect problem). We allow the number of latent vectors $K$ to grow at appropriate rates. We also develop a selection criterion when $K$ is unknown that provably does not under-select. We show that the theoretical guarantees hold with the selected number for $K$, and study the bias rates when $K$ is misspecified. With the new methods, we study peer effects in conflict and school climate perception using data on more than 7000 students from 23 schools.
JMLR Machine Learning · 发表于 2026
Pablo Barceló, Alexander Kozachinskiy, Tomasz Steifer
摘要
The notion of rank of a Boolean function has been a cornerstone in PAC learning, enabling quasipolynomial-time learning algorithms for polynomial-size decision trees. We present a novel characterization of rank, grounded in the well-known Transformer architecture. We show that the rank of a function $f$ corresponds to the minimum number of Chain of Thought (CoT) iterations required by a single-layer Transformer with hard attention to compute $f$. Based on this characterization, we establish tight bounds on the number of CoT iterations required for specific problems, showing that \(\ell\)-fold function composition necessitates exactly \(\ell\) CoT iterations. Furthermore, we analyze the problem of identifying the position of the \(k\)-th occurrence of 1 in a Boolean sequence, proving that it requires \(k\) CoT iterations. Finally, we introduce the notion of the multi-head rank that captures multi-head single-layer transformers, and perform the analysis of PAC-learnability of the classes of functions with bounded multi-head rank.
JMLR Other · 发表于 2026
Maolin Che, Yimin Wei, Hong Yan
摘要
In the framework of the FD (frequent directions) algorithm, we first develop two efficient algorithms for low-rank matrix approximations under the embedding matrices composed of the product of any SpEmb (sparse embedding) matrix and any standard Gaussian matrix, or any SpEmb matrix and any SRHT (subsampled randomized Hadamard transform) matrix. The theoretical results are also achieved based on the bounds of singular values of standard Gaussian matrices and the theoretical results for SpEmb and SRHT matrices. With a given Tucker-rank, we then obtain several efficient FD-based randomized variants of T-HOSVD (the truncated high-order singular value decomposition) and ST-HOSVD (sequentially T-HOSVD), which are two common algorithms for computing the approximate Tucker decomposition of any tensor with a given Tucker-rank. We also consider efficient FD-based randomized algorithms for computing the approximate TT (tensor-train) decomposition of any tensor with a given TT-rank. Finally, we illustrate the efficiency and accuracy of these algorithms using synthetic and real-world matrix (and tensor) data.
Biometrics Survival Analysis Nonparametric & Semiparametric Machine Learning · 发表于 2026,收录于 2026-08-17
Qiang Wu, Mingyue Du, Shuangge Ma, Xingqiu Zhao
摘要
For the analysis of interval-censored data, we propose a deep generalized accelerated hazards model. This model is designed to facilitate a detailed exploration of the relationship between various risk factors and the hazard associated with failure time. We develop a sieve maximum likelihood estimation procedure that combines deep neural networks and monotonic splines. By employing deep neural networks, we can effectively capture nonparametric effects, enabling a flexible and adaptive modeling approach for complex relationships. Under certain regularity conditions, we derive a nonasymptotic error bound for the resulting estimator and show that the finite-dimensional estimator is asymptotically normal and achieves the semiparametric efficiency. We conduct simulation studies to evaluate the finite-sample performance of the proposed approach. Furthermore, the proposed method is applied to the Atherosclerosis Risk in Communities study for practical illustration.
SiM Diagnostics & Prediction Machine Learning Networks & Graphs · 发表于 2026,收录于 2026-08-17
Qunqiang Feng, Taozheng Jia, Pan Liu, Ben Van Calster, Jialiang Li
摘要
Evaluating diagnostic accuracy for multi‐category outcomes remains a significant challenge, primarily due to computational limitations in existing performance metrics. The Polytomous Discrimination Index (PDI) has emerged as an order‐agnostic solution suitable for nominal classifications. However, its broader adoption has been constrained by the lack of efficient implementation, especially for its variance estimation, which typically would require computationally intensive bootstrapping procedures. In this work, we address this limitation by proposing a novel asymptotic variance estimator for the PDI. Our method integrates classical ‐statistic theory with recent advances in combinatorics, offering a scalable and theoretically grounded alternative. To assess the performance of the proposed approach, we conduct extensive simulation studies and observe remarkable gain in computing time. We further apply our method to a real‐world brain image analysis where deep neural networks are used as a diagnostic tool. We can efficiently report the accuracy of the neural networks with different depth specifications.
JMLR High-Dimensional Meta-Analysis & Data Integration · 发表于 2026
Yue Liu, Molei Liu, Zijian Guo, Tianxi Cai
摘要
Surrogate variables play an important role in various fields due to the scarcity or absence of gold-standard labels. We develop a novel approach named SASH for Surrogate-Assisted and data-Shielding High-dimensional integrative regression. It is a semi-supervised approach that efficiently leverages sizable unlabeled samples with error-prone surrogate outcomes from multiple local sites to improve model estimation using the small gold-labeled sample. To facilitate stable and efficient knowledge extraction from the surrogates, our method first obtains a preliminary supervised estimator, and then uses it to assist in training a regularized single-index model (SIM) for the surrogates. Interestingly, through a chain of convex and properly penalized sparse regressions that approximate the SIM loss using bias correction, our method avoids the problem of local minima in the SIM, and fully eliminates the impact of the preliminary estimator's excessive error. In addition, it protects individual-level information through the aggregation of summary statistics from local sites, leveraging a similar idea of bias-corrected approximation. Through simulation studies, we demonstrate that our method outperforms existing approaches. Finally, we apply our method to develop a genetic risk model for type 2 diabetes using large-scale data sets from UK and Mass General Brigham biobanks, where only a small fraction of subjects in one site are labeled through manual chart review.
JMLR Other · 发表于 2026,收录于 2026-09-10
Seong-ho Lee, Yanyuan Ma, Jiwei Zhao
摘要
(verbatim, from arXiv preprint arxiv.org/abs/2508.17780 — the JMLR page itself could not be re-fetched to confirm word-for-word identity): "In many real-world applications, researchers aim to deploy models trained in a source domain to a target domain, where obtaining labeled data is often expensive, time-consuming, or even infeasible. While most existing literature assumes that the labeled source data and the unlabeled target data follow the same distribution, distribution shifts are common in practice. This paper focuses on label shift and develops efficient inference procedures for general parameters characterizing the unlabeled target population. A central idea is to model the outcome density ratio between the labeled and unlabeled data. To this end, we propose a progressive estimation strategy that unfolds in three stages: an initial heuristic guess, a consistent estimation, and ultimately, an efficient estimation. This self-evolving process is novel in the statistical literature and of independent interest. We also highlight the connection between our approach and prediction-powered inference (PPI), which uses machine learning models to improve statistical inference in related settings. We rigorously establish the asymptotic properties of the proposed estimators and demonstrate their superior performance compared to existing methods. Through simulation studies and multiple real-world applications, we illustrate both the theoretical contributions and practical benefits of our approach.
JMLR Causal Inference · 发表于 2026,收录于 2026-11-06
Trinetri Ghosh, Jiawei Shan, Menggang Yu, Jiwei Zhao
摘要
In this paper, we consider estimation of the average treatment effect on the treated (ATT), an interpretable and relevant causal estimand to policymakers when treatment assignment is endogenous. By considering shadow variables that are unrelated to the treatment assignment but related to the outcomes of interest, we establish identification of the ATT. Then we focus on efficient estimation of the ATT by characterizing the geometric structure of the likelihood, deriving the semiparametric efficiency bound for ATT estimation and proposing an estimator that can achieve this bound. We rigorously establish the theoretical results of the proposed estimator. The finite sample performance of the proposed estimator is studied through comprehensive simulation studies as well as an application to our motivating study.
Econometrica Econometrics & Economics · 发表于 2026
Zhang Chen
摘要
I document that the right tail of the firm size distribution systematically thickens with economic development. To rationalize this fact, I develop a parsimonious idea search model in which both aggregate growth and the firm size distribution are endogenously determined. The model features an asymptotic balanced growth path along which Gibrat's law holds at each date, and the right tail of the firm size distribution thickens monotonically toward Zipf's law. The model also implies that policies favoring large firms can improve welfare by better utilizing the diffusion externalities arising from idea search.
Econometrica Other · 发表于 2026
Johannes Hörner, Anna Sanktjohanser
摘要
We consider a dynamic buyer–seller interaction. Instead of the buyer's valuation, it is the frequency with which he needs to trade that is the buyer's private information. The difference matters. With commitment, full surplus extraction is possible, for instance via limited‐time offers. Without commitment, ratcheting is mitigated, as not buying is not necessarily a sign of strength. Because time is informative, the seller learns and may adjust her behavior over time. When finding a suitable alternative seller is easy, she starts with a pooling offer before permanently switching to a screening offer. When finding a suitable alternative seller takes time, she starts with a pooling offer before occasionally experimenting with separating offers.
Econometrica Econometrics & Economics · 发表于 2026
Gabriel Carroll
摘要
Consider a repeated interaction where it is unknown which of various stage games will be played each period. This framework separates the basic logic of intertemporal incentives from the requirement that any given strategy profile yields a well‐defined payoff vector. A natural solution concept is ex post perfect equilibrium: strategies must form a subgame‐perfect equilibrium for any realization of the sequence of stage games. When there is one long‐run player and others are short‐run, and public randomization is available, we can adapt the standard recursive approach to determine the maximum feasible gap between reward and punishment for the long‐run player. This allows us to identify which actions can be played in equilibrium and, assuming perfect monitoring, to fully characterize what outcome paths can arise. With multiple long‐run players or no public randomization, the approach fails; a diagnostic of this failure is that optimal penal codes may no longer exist.
JMLR Meta-Analysis & Data Integration · 发表于 2026
Tao Wang, Weng Kee Wong
摘要
This paper develops a general framework for doubly debiased robust subsampling for transfer learning. The setting arises when massive source datasets are computationally infeasible to use in full, while naive or heuristic subsampling leads to biased estimators that further inherit transfer bias under source-target distributional shifts. We resolve these challenges through two complementary debiasing mechanisms. Inverse probability weighting removes subsampling bias by ensuring that subsample-based estimators represent the full source distribution, while a target-based one-step refinement recenters estimators towards the target distribution, thereby mitigating transfer bias. These corrections are embedded within a distributionally robust optimization design that simultaneously controls worst-case target risk and enforces source-target alignment through maximum mean discrepancy. To optimize subsampling distributions, we propose a scalarized particle swarm algorithm that efficiently explores the robustness-alignment frontier by adjusting a single tuning parameter. We establish theoretical properties, including asymptotic normality, generalization bounds, oracle inequalities, and minimax optimality under distributional uncertainty. Simulation studies and empirical applications in text sentiment and image recognition demonstrate that the proposed method consistently improves prediction accuracy and robustness compared with uniform subsampling, target-only training, and alignment-only approaches, and that both debiasing mechanisms are essential for reliable transfer.
Econometrica Testing & Uncertainty Spatial & Time Series · 发表于 2026
José Luis Montiel Olea, Mikkel Plagborg-Møller, Eric Qian, Christian K. Wolf
摘要
We consider impulse response inference in a locally misspecified vector autoregression (VAR) model. The conventional local projection (LP) confidence interval has correct coverage even when the misspecification is so large that it can be detected with probability approaching 1. This result follows from a “double robustness” property analogous to that of popular partially linear regression estimators. By contrast, the conventional VAR confidence interval with short‐to‐moderate lag length can severely undercover for misspecification that is small, difficult to detect statistically, and cannot be ruled out based on economic theory. The VAR confidence interval has robust coverage if, and only if, the lag length is so large that the interval is as wide as the LP interval.
JMLR Other · 发表于 2026
Doudou Zhou, Mengyan Li, Yun Wang, Tianxi Cai, Molei Liu
摘要
Domain adaptation enables generalizable and efficient data-driven research. However, existing work has largely focused on domain adaptation for some intrinsically homogeneous target cohort, overlooking inherent heterogeneity within the target, which can exacerbate biases and unfairness in the presence of subgroups with imbalanced sample sizes. We develop a novel domain adaptation framework that addresses a more complicated target dataset that consists of heterogeneous and data-sparse subgroups and lacks gold-standard label observations. Our method simultaneously handles high-dimensionality, covariate shift, and outcome model heterogeneity by combining a model-assisted debiasing step used for covariate shift correction with an adaptive knowledge-guided sparsification procedure used to mitigate the issue of sample disparity. We also introduce a new model selection strategy to avoid negative knowledge transfer in the absence of labels in the target data. Our method is theoretically justified for being robust to nuisance model misspecification and adaptive to heterogeneity between the subgroups. Numerical experiments and two real-world applications, including genetic risk modeling of type 2 diabetes and prediction of mutation-induced protein stability changes, demonstrate the practical advantages of our method.
JMLR Other · 发表于 2026
Yexiong Lin, Yu Yao, Yuxuan Du, Jun Yu, Bo Han, Mingming Gong, Tongliang Liu
摘要
Statistically consistent algorithms have been widely employed for dealing with noisy labels. Their objective functions are designed so that minimizing the expected risk on noisy data leads to the same minimizer as minimizing the expected risk on clean data. From the weak law of large numbers, penalizing the variance of losses would reduce the discrepancy between the average loss and the expected risk on the clean data when there is a finite training sample, and the estimation error in the model's parameters can be reduced. Interestingly, we found that the variance of losses needs to be encouraged for label-noise learning. Specifically, encouraging a large variance of losses would boost the memorization effect and reduce the harmfulness of incorrect labels. Regularizers can be easily designed to encourage a large variance of losses and be plugged into many existing algorithms. Empirically, the proposed method by encouraging a large variance of losses could improve the generalization ability of baselines on both synthetic and real-world datasets.
JMLR Causal Inference Machine Learning · 发表于 2026,收录于 2026-11-06
Chenxi Liu, Yongqiang Chen, Tongliang Liu, Mingming Gong, James Cheng, Bo Han, Kun Zhang
摘要
Revealing hidden causal variables alongside the underlying causal mechanisms is essential to the development of science. Despite the progress in the past decades, existing practice in causal discovery (CD) heavily relies on high-quality measured variables, which are usually given by human experts. In fact, the lack of well-defined high-level variables behind unstructured data has been a longstanding roadblock to a broader real-world application of CD. This procedure can naturally benefit from an automated process that can suggest potential hidden variables in the system. Interestingly, Large language models (LLMs) are trained on massive observations of the world and have demonstrated great capability in processing unstructured data. To leverage the power of LLMs, we develop a new framework termed Causal representation Assistant (COAT) that incorporates the rich world knowledge of LLMs to propose useful measured variables for CD with respect to high-value target variables on their paired unstructured data. Instead of directly inferring causality with LLMs, COAT constructs feedback from intermediate CD results to LLMs to refine the proposed variables. Given the target variable and the paired unstructured data, we first develop COAT-MB that leverages the predictivity of the proposed variables to iteratively uncover the Markov Blanket of the target variable. Built upon COAT-MB, COAT-PAG further extends to uncover a more complete causal graph, i.e., Partial Ancestral Graph, by iterating over the target variables and actively seeking new high-level variables. Moreover, the reliable CD capabilities of COAT also extend the debiased causal inference to unstructured data by discovering an adjustment set. We establish theoretical guarantees for the CD results and verify that COAT can effectively and reliably make the most use of LLM knowledge for CD across 7 realistic benchmarks and real-world case studies, including analysis of human reviews and diagnosis of neuropathic and brain tumors.
JMLR Other · 发表于 2026
Kevin Miller, Ryan Murray
摘要
This work introduces Dirichlet Active Learning (DiAL), a Bayesian-inspired approach to the design of active learning algorithms. Our framework models feature-conditional class probabilities as a Dirichlet random field and lends observational strength between similar features in order to calibrate the random field. This random field can then be utilized in learning tasks: in particular, we can use current estimates of mean and variance to conduct classification and active learning in the context where labeled data is scarce. We demonstrate the applicability of this model to low-label rate graph learning by constructing “propagation operators” based upon the graph Laplacian, and offer computational studies demonstrating the method's competitiveness with the state of the art. Finally, we provide rigorous guarantees regarding the ability of this approach to ensure both exploration and exploitation, expressed respectively in terms of cluster exploration and increased attention to decision boundaries.
JMLR Other · 发表于 2026
Dingcheng Luo, Thomas O'Leary-Roseberry, Peng Chen, Omar Ghattas
摘要
We study the derivative-informed learning of nonlinear operators between infinite-dimensional Hilbert spaces. Such operators can arise as solution maps of partial differential equations, and their approximation by accurate surrogate models can accelerate simulation-intensive tasks of scientific and engineering interest, including inference, control, and uncertainty quantification. Since efficiently performing such tasks often requires an accurate representation of the operator's derivatives, we analyze the approximation capabilities of neural operators measured by Sobolev norms over infinite-dimensional Gaussian input measures. We focus on the reduced basis neural operator, which employs linear encoders/decoders defined on dominant input/output subspaces. We study two methods for generating the subspaces: principal component analysis (PCA) and derivative-informed subspaces (DIS). These use the dominant eigenvectors of the covariance of the data or the derivatives as bases for the subspaces, respectively. We derive bounds for errors arising from both dimension reduction and latent neural network approximations, including sampling errors associated with the empirical estimation of the PCA/DIS subspaces. Our analysis is validated through numerical experiments, which demonstrate that subspaces informed by the underlying operator (DIS or output PCA) yield smaller generalization errors in the Sobolev norm, while input PCA may underperform unless ranks and training sample sizes are sufficiently large.
JMLR Other · 发表于 2026
Kaveh Alim, Hao Wang, Ojas Gulati, Akash Srivastava, Navid Azizan
摘要
Existing differentially private (DP) synthetic data generation mechanisms typically assume a single-source table. In practice, data is often distributed across multiple tables with relationships across tables. In this paper, we introduce the first-of-its-kind algorithm that can be combined with any existing DP mechanisms to generate synthetic relational databases. Our algorithm iteratively refines the relationship between individual synthetic tables to minimize their approximation errors in terms of low-order marginal distributions while maintaining referential integrity. This algorithm eliminates the need to flatten a relational database into a master table (saving space), operates efficiently (saving time), and scales effectively to high-dimensional data. We provide both DP and theoretical utility guarantees for our algorithm. Through numerical experiments on real-world datasets, we demonstrate the effectiveness of our method in preserving fidelity to the original data.
JMLR High-Dimensional Testing & Uncertainty · 发表于 2026
Zhanrui Cai, Sai Li, Xintao Xia, Linjun Zhang
摘要
This paper proposes new methodologies for conducting practical differentially private (DP) estimation and inference in high-dimensional linear regression. We first introduce a DP Bayesian Information Criterion (DP-BIC) for selecting the unknown sparsity parameter in differentially private sparse linear regression (DP-SLR), eliminating the need for prior knowledge of model sparsity, which is a requisite in the existing literature. Next, we develop the DP debiased algorithm that enables privacy-preserving inference on a particular subset of regression parameters. Our proposed method enables privacy-preserving inference on the regression parameters by leveraging the inherent sparsity of high-dimensional linear regression models. Additionally, we address private feature selection by considering multiple testing in high-dimensional linear regression by introducing a DP multiple testing procedure that controls the false discovery rate (FDR). This allows for accurate and privacy-preserving identification of significant predictors in the regression model. Through extensive simulations and real data analyses, we demonstrate the effectiveness of our proposed methods in conducting inference for high-dimensional linear models while safeguarding privacy and controlling the FDR.
JMLR Other · 发表于 2026
Achraf Azize, Marc Jourdan, Aymen Al Marjani, Debabrota Basu
摘要
Best Arm Identification (BAI) problems are progressively used for data-sensitive applications, such as designing adaptive clinical trials, tuning hyper-parameters, and conducting user studies. Motivated by the data privacy concerns invoked by these applications, we study the problem of BAI with fixed confidence in both the local and central models, i.e. under $\epsilon$-local and $\epsilon$-global Differential Privacy (DP). First, to quantify the cost of privacy, we derive lower bounds on the sample complexity of any $\delta$-correct BAI algorithm satisfying$\epsilon$-global DP or $\epsilon$-local DP. Our lower bounds suggest the existence of two privacy regimes. In the high-privacy regime, the hardness depends on a coupled effect of privacy and novel information-theoretic quantities involving the Total Variation distance. In the low-privacy regime, the lower bounds reduce to the non-private lower bounds. We propose $\epsilon$-local DP and $\epsilon$-global DP variants of a Top Two algorithm, namely CTB-TT and AdaP-TT$\star$, respectively. For $\epsilon$-local DP, CTB-TT is asymptotically optimal by plugging in a private estimator of the means based on Randomised Response. For $\epsilon$-global DP, our private estimator of the mean runs in arm-dependent adaptive episodes and adds Laplace noise to ensure a good privacy-utility trade-off. By adapting the transportation costs, the expected sample complexity of AdaP-TT$\star$ reaches the asymptotic lower bound in the asymptotic high-privacy regime, up to a small multiplicative constant.
JMLR Other · 发表于 2026
Vincent Divol, Solenne Gaucher
摘要
This paper explores the theoretical foundations of fair regression under the constraint of demographic parity within the unawareness framework, where disparate treatment is prohibited, extending existing results where such treatment is permitted. Specifically, we aim to characterize the optimal fair regression function when minimizing the quadratic loss. Our results reveal that this function is given by the solution to a barycenter problem with optimal transport costs. Additionally, we study the connection between optimal fair cost-sensitive classification, and optimal fair regression. We demonstrate that nestedness of the decision sets of the classifiers is both necessary and sufficient to establish a form of equivalence between classification and regression. Under this nestedness assumption, the optimal classifiers can be derived by applying thresholds to the optimal fair regression function; conversely, the optimal fair regression function is characterized by the family of cost-sensitive classifiers.
JMLR Nonparametric & Semiparametric Causal Inference · 发表于 2026
Marco Simnacher, Xiangnan Xu, Hani Park, Christoph Lippert, Sonja Greven
摘要
Conditional independence tests (CITs) test for conditional dependence between random variables given a vector of conditioning or confounder variables. As existing CITs are limited in their applicability to complex, high-dimensional variables such as images, we introduce deep nonparametric CITs (DNCITs). The DNCITs combine embedding maps, which extract feature representations of high-dimensional variables, with nonparametric CITs applicable to these feature representations. For the embedding maps, we derive general properties on their parameter estimators to obtain valid DNCITs and show that these properties include embedding maps learned through (conditional) unsupervised or transfer learning. For the nonparametric CITs, appropriate tests are selected and adapted to be applicable to feature representations. Through simulations, we investigate the performance of the DNCITs for different embedding maps and nonparametric CITs under varying confounder dimensions and confounder relationships. We apply the DNCITs to brain MRI scans and behavioral traits, given confounders, of healthy individuals from the UK Biobank, confirming null results from a number of ambiguous personality neuroscience studies, now with a larger data set and with our more powerful tests. In addition, in a confounder control study, we apply the DNCITs to brain MRI scans and a confounder set to test for sufficient confounder control. We provide an R package implementing the proposed DNCITs.
JMLR Machine Learning Networks & Graphs Survival Analysis · 发表于 2026,收录于 2026-11-06
Dongxiao Han, Zan Li, Xinyuan Song, Donglin Zeng, Liuquan Sun
摘要
In this article, we propose a methodology for learning a risk score based on a general loss function, in which the score function is parameterized by a rectified linear unit feedforward deep neural network. The non-asymptotic error bound of the loss function is established and three examples are provided to illustrate the proposed method. The finite sample behavior of the proposed method is evaluated through simulation studies and an application to a SEER colon cancer data is provided.
JMLR Networks & Graphs · 发表于 2026
Myeonghun Yu, Kean Ming Tan, Huixia Judy Wang, Wen-Xin Zhou
摘要
Expected shortfall (ES), also known as conditional value-at-risk, is a widely recognized risk measure that complements value-at-risk by capturing tail-related risks more effectively. Compared with quantile regression, which has been extensively developed and applied across disciplines, ES regression remains in its early stage, partly because the traditional empirical risk minimization framework is not directly applicable. In this paper, we develop a nonparametric framework for expected shortfall regression based on a two-step approach that treats the conditional quantile function as a nuisance parameter. Leveraging the representational power of deep neural networks, we construct a two-step ES estimator using feedforward ReLU networks, which can alleviate the curse of dimensionality when the underlying functions possess hierarchical composition structures. However, ES estimation is inherently sensitive to heavy-tailed response or error distributions. To address this challenge, we integrate a properly tuned Huber loss into the neural network training, yielding a robust deep ES estimator that is provably resistant to heavy-tailedness in a non-asymptotic sense and first-order insensitive to quantile estimation errors in the first stage. Comprehensive simulation studies and an empirical analysis of the effect of El Niño on extreme precipitation illustrate the accuracy and robustness of the proposed method.
JMLR High-Dimensional Testing & Uncertainty · 发表于 2026
Zijian Guo, Wei Yuan, Cunhui Zhang
摘要
Additive models play an essential role in studying non-linear relationships. Despite many recent advances in estimation, there is a lack of methods and theories for inference in high-dimensional additive models, including confidence interval construction and hypothesis testing. Motivated by inference for non-linear treatment effects, we consider the high-dimensional additive model and make inferences for the function derivative. We propose a novel decorrelated local linear estimator and establish its asymptotic normality. The main novelty is the construction of the decorrelation weights, which is instrumental in reducing the error inherited from estimating the nuisance functions in the high-dimensional additive model. We construct the confidence interval for the function derivative and conduct the related hypothesis testing. We demonstrate our proposed method over large-scale simulation studies and apply it to identify non-linear effects in the motif regression problem. Our proposed method is implemented in the R package DLL available from CRAN.
JMLR Nonparametric & Semiparametric Latent Variables & Mixtures · 发表于 2026
Balabdaoui, Fadoua, Di Noia, Antonio, Durot, Cécile
摘要
Unlinked regression, in which covariates and responses are observed separately without known correspondence, has recently gained increasing attention. Deconvolution, on the other hand, is a fundamental and challenging problem in nonparametric statistics with the aim of estimating the distribution of a latent random variable $Z$ based on observations contaminated by some additive noise. The complexity of this task is heavily influenced by the smoothness of the noise distribution and often leads to slow estimation rates. In this paper, we combine the recent unlinked linear regression problem with the classical deconvolution framework. Specifically, we study nonparametric deconvolution under the assumption that $Z$ is a linear function of an observable multidimensional covariate. This structural constraint allows us to introduce a nonparametric estimator of the distribution of $Z$ which achieves the parametric rate of convergence in the Wasserstein distance of order 1, where the smoothness of the noise does not affect the rate. Furthermore, we introduce nonparametric estimators for the unconditional density of $Z$ and the conditional density of $Z$ given an observed response. This allows us to study the problem of estimating the value of the latent linear predictor, whose link to the observed response is not accessible. Through several simulations, we illustrate the fast convergence rate of our deconvolution estimator and the performance of the proposed conditional estimators of the latent predictor in different simulation scenarios.
JMLR Optimization & Learning Theory · 发表于 2026
TIanyu Cao, Xiaokai Chen, Gesualdo Scutari
摘要
We study decentralized optimization over a network of agents, modeled as an undirected graph and operating without a central server. The objective is to minimize a composite function $f+r$, where $f$ is a (strongly) convex function representing the average of the agents' losses, and $r$ is a convex, extended-value function (regularizer). We introduce DCatalyst, a unified black-box framework that injects Nesterov-type acceleration into decentralized optimization algorithms. At its core, DCatalyst is an inexact, momentum-accelerated proximal scheme (outer loop) that seamlessly wraps around a given decentralized method (inner loop). We show that DCatalyst attains optimal (up to logarithmic factors) communication and computational complexity across a broad family of decentralized algorithms and problem instances. In particular, it delivers accelerated rates for problem classes that previously lacked accelerated decentralized methods, thereby broadening the effectiveness of decentralized methods. On the technical side, our framework introduces inexact estimating sequences--an extension of Nesterov's classical estimating sequences, tailored to decentralized, composite optimization. This construction systematically accommodates consensus errors and inexact solutions of local subproblems, addressing challenges that existing estimating-sequence-based analyses cannot handle while retaining a black-box, plug-and-play character.
Econometrica Other · 发表于 2026,收录于 2026-09-24
Sergei Guriev, Emeric Henry, Théo Marquis, Ekaterina Zhuravskaya
摘要
(from a RePEc/EconPapers mirror of the December 2023 CEPR/SSRN working-paper version; the econometricsociety.org paraphrase of the final version references a second experiment during "the 2022 and 2024 U.S. elections," while this text only covers the 2022 midterms, so the published abstract has likely been updated and may differ from the text below) "We develop a comprehensive framework to assess policy measures aimed at curbing false news dissemination on social media. A randomized experiment on Twitter during the 2022 U.S. mid-term elections evaluates such policies as priming the awareness of misinformation, fact-checking, confirmation clicks, and prompting careful consideration of content. Priming is the most effective policy in reducing sharing of false news while increasing sharing of true content. A model of sharing decisions, motivated by persuasion, partisan signaling, and reputation concerns, predicts that policies affect sharing through three channels: (i) updating perceived veracity and partisanship of content, (ii) raising the salience of reputation, and (iii) increasing sharing frictions. Structural estimation shows that all policies impact sharing via the salience of reputation and cost of friction. Affecting perceived veracity plays a negligible role as a mechanism in all policies, including fact-checking. The priming intervention performs best in enhancing reputation salience with minimal added friction.
JMLR Bayesian Methods Genetics & Omics · 发表于 2026
Huizi Zhang, Sara Wade, Natalia Bochkina
摘要
Bayesian hierarchical modeling is a natural framework to effectively integrate data and borrow information across groups. In this paper, we address problems related to density estimation and identifying clusters across related groups, by proposing a hierarchical Bayesian approach that incorporates additional covariate information. To achieve flexibility, our approach builds on ideas from Bayesian nonparametrics, combining the hierarchical Dirichlet process with dependent Dirichlet processes. The proposed model is widely applicable, accommodating multiple and mixed covariate types through appropriate kernel functions as well as different output types through suitable component-specific likelihoods. This extends our ability to discern the relationship between covariates and clusters, while also effectively borrowing information and quantifying differences across groups. By employing a data augmentation trick, we are able to tackle the intractable normalized weights and construct a Markov chain Monte Carlo algorithm for posterior inference. The proposed method is illustrated on simulated data and two real data sets on single-cell RNA sequencing (scRNA-seq) and calcium imaging. For scRNA-seq data, we show that the incorporation of cell dynamics facilitates the discovery of additional cell subgroups. On calcium imaging data, our method identifies interpretable clusters of time frames with similar neural activity, aligning with the observed behavior of the animal.
JMLR Other · 发表于 2026
Laura Iacovissi, Nan Lu, Robert C. Williamson
摘要
Corruption is notoriously widespread in data collection. Despite extensive research, the existing literature predominantly focuses on specific settings and learning scenarios, lacking a unified view of corruption modelization and mitigation. In this work, we develop a general theory of corruption, which incorporates all modifications to a supervised learning problem, including changes in model class and loss. Focusing on changes to the underlying probability distributions via Markov kernels, our approach leads to three novel opportunities. First, it enables the construction of a novel, provably exhaustive corruption framework, distinguishing among different corruption types. This serves to unify existing models and establish a consistent nomenclature. Second, it facilitates a systematic analysis of corruption consequences on learning tasks, by considering Bayes risks in the clean and corrupted scenarios. Notably, while label corruptions affect only the loss function, attribute corruptions additionally influence the hypothesis class. Third, building upon these results, we investigate mitigations for various corruption types. We expand existing loss-correction methods for label corruption to handle dependent corruption types. Our findings highlight the necessity to generalize this classical corruption-corrected learning framework to a new paradigm with weaker requirements to encompass more corruption types. We provide such a paradigm as well as loss correction formulas in the attribute and joint corruption cases.
Econometrica Econometrics & Economics · 发表于 2026
Allan Hsiao
摘要
Weak environmental regulation has global consequences. When domestic regulation fails, the international community can target emitters with trade policy. I develop a dynamic empirical framework for evaluating trade policy as a substitute for domestic regulation, and I apply the framework to the market for palm oil, a major driver of deforestation and global CO 2 emissions. Relative to business as usual, a domestic production tax of 50% reduces CO 2 emissions by 7.4 Gt from 1988 to 2016, amounting to 0.26 Gt annually. Coordinated, committed import tariffs of similar magnitude reduce emissions by 5.4 Gt over the same period. The cost of these import tariffs is only $15 per ton of CO 2 , even accounting for compensating transfers that recognize welfare losses for producing countries. Without coordination and commitment, import tariffs have more limited effects. Alternative policies include domestic export taxes, which are fiscally appealing independent of emission concerns, and a carbon border adjustment mechanism, which encourages domestic regulation.
JMLR Networks & Graphs Machine Learning Optimization & Learning Theory · 发表于 2026,收录于 2026-11-06
Sourav Chatterjee
摘要
We give a simple local Polyak--Lojasiewicz (PL) criterion that guarantees linear (exponential) convergence of gradient flow and gradient descent to a zero-loss solution of a nonnegative objective. We then verify this criterion for the squared training loss of a feedforward neural network with smooth, strictly increasing activation functions, in a regime that is complementary to the usual over-parameterized analyses: the network width and depth are fixed, while the input data vectors are assumed to be linearly independent (in particular, the ambient input dimension is at least the number of data points). A notable feature of the verification is that it is constructive: it leads to a simple "positive'" initialization (zero first-layer weights, strictly positive hidden-layer weights, and sufficiently large output-layer weights) under which gradient descent provably converges to an interpolating global minimizer of the training loss. We also discuss a probabilistic corollary for random initializations, clarify its dependence on the probability of the required initialization event, and provide numerical experiments showing that this theory-guided initialization can substantially accelerate optimization relative to standard random initializations at the same width.