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
2025 年的期刊文章 · 第 37 页
共 1491 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2025 到 2025。
按期刊、按主题看 →
JMLR Other · 发表于 2025
Yujin Jeong, Dominik Rothenhäusler
摘要
How can we draw trustworthy scientific conclusions? One criterion is that a study can be replicated by independent teams. While replication is critically important, it is arguably insufficient. If a study is biased for some reason and other studies recapitulate the approach then findings might be consistently incorrect. It has been argued that trustworthy scientific conclusions require disparate sources of evidence. However, different methods might have shared biases, making it difficult to judge the trustworthiness of a result. We formalize this issue by introducing a "distributional uncertainty model", wherein dense distributional shifts emerge as the superposition of numerous small random changes. The distributional perturbation model arises under a symmetry assumption on distributional shifts and is strictly weaker than assuming that the data is i.i.d. from the target distribution. We show that a stability analysis on a single data set allows us to construct confidence intervals that account for both sampling uncertainty and distributional uncertainty.
JMLR Testing & Uncertainty Diagnostics & Prediction · 发表于 2025
Yajie Bao, Yuyang Huo, Haojie Ren, Changliang Zou
摘要
This paper studies the problem of post-selection predictive inference in an online fashion. To avoid devoting resources to unimportant units, a preliminary selection of the current individual before reporting its prediction interval is common and meaningful in online predictive tasks. As a result of the temporal multiplicity introduced by the online selection process, it becomes essential to control the real-time false coverage-statement rate (FCR) which measures the overall miscoverage level. We develop a general framework named CAP (Calibration after Adaptive Pick), which performs an adaptive pick rule on historical data to construct a calibration set if the current individual is selected. This is followed by the output of a conformal prediction interval for the unobserved label. We present a series of tractable procedures for the construction of the calibration set for various popular online selection rules. It has been demonstrated that CAP achieves an exact selection-conditional coverage guarantee in the finite-sample and distribution-free regimes. In order to address the issue of the distribution shift in online data, we also embed CAP into some recent dynamic conformal prediction algorithms and prove that the proposed method can deliver long-run FCR control. Numerical results on both synthetic and real data corroborate that CAP can effectively control FCR around the target level and yield narrower prediction intervals than existing baselines across various settings.
JMLR Causal Inference Machine Learning · 发表于 2025
Maximilian Kertel, Nadja Klein
摘要
We present a boosting-based method to learn additive Structural Equation Models (SEMs) from observational data, with a focus on the theoretical aspects of determining the causal order among variables. We introduce a family of score functions based on arbitrary regression techniques, for which we establish sufficient conditions that guarantee consistent identification of the true causal ordering. Our analysis reveals that boosting with early stopping meets these criteria and thus offers a consistent score function for causal orderings. To address the challenges posed by high-dimensional data sets, we adapt our approach through a component-wise gradient descent in the space of additive SEMs. Our simulation study supports the theoretical findings in low-dimensional settings and demonstrates that our high-dimensional adaptation is competitive with state-of-the-art methods. In addition, it exhibits robustness with respect to the choice of hyperparameters, thereby simplifying the tuning process.
JMLR Bayesian Methods Optimization & Learning Theory · 发表于 2025
Johannes P. Dürholt, Thomas S. Asche, Johanna Kleinekorte, Gabriel Mancino-Ball, Benjamin Schiller, Simon Sung, Julian Keupp, Aaron Osburg, Toby Boyne, Ruth Misener, Rosona Eldred, Chrysoula Kappatou, Robert M. Lee, Dominik Linzner, Wagner Steuer Costa, David Walz, Niklas Wulkow, Behrang Shafei
摘要
Our open-source Python package BoFire combines Bayesian Optimization (BO) with other design of experiments (DoE) strategies focusing on developing and optimizing new chemistry. Previous BO implementations, for example as they exist in the literature or software, require substantial adaptation for effective real-world deployment in chemical industry. BoFire provides a rich feature-set with extensive configurability and realizes our vision of fast-tracking research contributions into industrial use via maintainable open-source software. Owing to quality-of-life features like JSON-serializability of problem formulations, BoFire enables seamless integration of BO into RESTful APIs, a common architecture component for both self-driving laboratories and human-in-the-loop setups. This paper discusses the differences between BoFire and other BO implementations and outlines ways that BO research needs to be adapted for real-world use in a chemistry setting.
JMLR Machine Learning · 发表于 2025
Hongyu Wang, Shuming Ma, Lingxiao Ma, Lei Wang, Wenhui Wang, Li Dong, Shaohan Huang, Huaijie Wang, Jilong Xue, Ruiping Wang, Yi Wu, Furu Wei
摘要
The increasing size of large language models (LLMs) has posed challenges for deployment and raised concerns about environmental impact due to high energy consumption. Previous research typically applies quantization after pre-training. While these methods avoid the need for model retraining, they often cause notable accuracy loss at extremely low bit-widths. In this work, we explore the feasibility and scalability of 1-bit pre-training. We introduce BitNet b1 and BitNet b1.58, the scalable and stable 1-bit Transformer architecture designed for LLMs. Specifically, we introduce BitLinear as a drop-in replacement of the nn.Linear layer in order to train 1-bit weights from scratch. Experimental results show that BitNet b1 achieves competitive performance, compared to state-of-the-art 8-bit quantization methods and FP16 Transformer baselines. With the ternary weight, BitNet b1.58 matches the half-precision Transformer LLM with the same model size and training tokens in terms of both perplexity and end-task performance, while being significantly more cost-effective in terms of latency, memory, throughput, and energy consumption. More profoundly, BitNet defines a new scaling law and recipe for training new generations of LLMs that are both high-performance and cost-effective. It enables a new computation paradigm and opens the door for designing specific hardware optimized for 1-bit LLMs.
JMLR Nonparametric & Semiparametric · 发表于 2025
Alan N. Amin, Debora S. Marks, Eli N. Weinstein
摘要
Applying machine learning to biological sequences---DNA, RNA and protein---has enormous potential to advance human health and environmental sustainability. To support such high-stakes applications, it is important to develop models and evaluations that not only capture underlying biology, but also have theoretical guarantees of reliability and performance. In this article, we analyze kernel methods for biological sequences, including both hand-crafted kernels and deep neural network-based kernels. We show that popular biological kernels can severely fail at learning functions or distinguishing distributions. We then develop modified kernels that (1) are universal, characteristic, and metrize the space of distributions, and (2) preserve the underlying biological inductive biases and domain knowledge embedded in the original kernel. Our results rest on novel proof techniques for kernels that handle the structure of biological sequence space--discrete, variable length sequences--and biological notions of sequence similarity. We illustrate our theoretical results in simulation and on real biological data sets.
JMLR Other · 发表于 2025
Tom Heskes
摘要
Bias-variance decompositions are widely used to understand the generalization performance of machine learning models. While the squared error loss permits a straightforward decomposition, other loss functions - such as zero-one loss or $L_1$ loss - either fail to sum bias and variance to the expected loss or rely on definitions that lack the essential properties of meaningful bias and variance. Recent research has shown that clean decompositions can be achieved for the broader class of Bregman divergences, with the cross-entropy loss as a special case. However, the necessary and sufficient conditions for these decompositions remain an open question. In this paper, we address this question by studying continuous, nonnegative loss functions that satisfy the identity of indiscernibles (zero loss if and only if the two arguments are identical). We prove that so-called $g$-Bregman or rho-tau divergences are the only such loss functions that have a clean bias-variance decomposition. A $g$-Bregman divergence can be transformed into a standard Bregman divergence through an invertible change of variables. This makes the squared Euclidean distance, up to such a variable transformation, the only symmetric loss function with a clean bias-variance decomposition. Consequently, common metrics such as zero-one and $L_1$ losses cannot admit a clean bias-variance decomposition, explaining why previous attempts have failed. We also examine the impact of relaxing the restrictions on the loss functions and how this affects our results.
JMLR Survey & Sampling · 发表于 2025
Jordan Awan, Adam Edwards, Paul Bartholomew, Andrew Sillers
摘要
In differential privacy (DP) mechanisms, it can be beneficial to release "redundant" outputs, where some quantities can be estimated in multiple ways by combining different privatized values. Indeed, the DP 2020 Decennial Census products published by the U.S. Census Bureau consist of such redundant noisy counts. When redundancy is present, the DP output can be improved by enforcing self-consistency (i.e., estimators obtained using different noisy counts result in the same value), and we show that the minimum variance processing is a linear projection. However, standard projection algorithms require excessive computation and memory, making them impractical for large-scale applications such as the Decennial Census. We propose the Scalable Efficient Algorithm for Best Linear Unbiased Estimate (SEA BLUE), based on a two-step process of aggregation and differencing that 1) enforces self-consistency through a linear and unbiased procedure, 2) is computationally and memory efficient, 3) achieves the minimum variance solution under certain structural assumptions, and 4) is empirically shown to be robust to violations of these structural assumptions. We propose three methods of calculating confidence intervals from our estimates, under various assumptions. Finally, we apply SEA BLUE to two 2010 Census demonstration products, illustrating its scalability and validity.
JMLR High-Dimensional Latent Variables & Mixtures Bayesian Methods · 发表于 2025
Dapeng Yao, Fangzheng Xie, Yanxun Xu
摘要
We study the sparse high-dimensional Gaussian mixture model when the number of clusters is allowed to grow with the sample size. A minimax lower bound for parameter estimation is established, and we show that a constrained maximum likelihood estimator achieves the minimax lower bound. However, this optimization-based estimator is computationally intractable because the objective function is highly nonconvex and the feasible set involves discrete structures. To address the computational challenge, we propose a computationally tractable Bayesian approach to estimate high-dimensional Gaussian mixtures whose cluster centers exhibit sparsity using a continuous spike-and-slab prior. We further prove that the posterior contraction rate of the proposed Bayesian method is minimax optimal. The mis- clustering rate is obtained as a by-product using tools from matrix perturbation theory. The proposed Bayesian sparse Gaussian mixture model does not require pre-specifying the number of clusters, which can be adaptively estimated. The validity and usefulness of the proposed method is demonstrated through simulation studies and the analysis of a real-world single-cell RNA sequencing data set.
JMLR Bayesian Methods Diagnostics & Prediction · 发表于 2025
Joshua J. Bon, David J. Warne, David J. Nott, Christopher Drovandi
摘要
Scientists continue to develop increasingly complex mechanistic models to reflect their knowledge more realistically. Statistical inference using these models can be challenging since the corresponding likelihood function is often intractable and model simulation may be computationally burdensome. Fortunately, in many of these situations it is possible to adopt a surrogate model or approximate likelihood function. It may be convenient to conduct Bayesian inference directly with a surrogate, but this can result in a posterior with poor uncertainty quantification. In this paper, we propose a new method for adjusting approximate posterior samples to reduce bias and improve posterior coverage properties. We do this by optimizing a transformation of the approximate posterior, the result of which maximizes a scoring rule. Our approach requires only a (fixed) small number of complex model simulations and is numerically stable. We develop supporting theory for our method and demonstrate beneficial corrections to approximate posteriors across several examples of increasing complexity.
JMLR Bayesian Methods Networks & Graphs Machine Learning · 发表于 2025
Ben Wu, Keru Wu, Jian Kang
摘要
Deep neural networks (DNN) have been widely used in scalar-on-image regression to predict an outcome variable from imaging predictors. However, training DNN typically requires large sample sizes for accurate prediction, and the resulting models often lack interpretability. In this work, we propose a novel Bayesian nonlinear scalar-on-image regression framework with a spatially varying single-layer neural network (SV-NN) prior. The SV-NN is constructed using a single hidden layer neural network with its weights generated by the soft-thresholded Gaussian process. Our framework enables the selection of interpretable image regions while achieving high prediction accuracy with limited training samples. The SV-NN offers large prior support for the imaging effect function, facilitating efficient posterior inference on image region selection and automatic network structures determination. We establish the posterior consistency for model parameters and selection consistency for image regions when the number of voxels/pixels grows much faster than the sample size. To ensure computational efficiency, we develop a stochastic gradient Langevin dynamics (SGLD) algorithm for posterior inference. We evaluate our method through extensive comparisons with state-of-the-art deep learning approaches, analyzing multiple real datasets, including task fMRI data from the Adolescent Brain Cognitive Development (ABCD) study.
JMLR Bayesian Methods · 发表于 2025
Didong Li, Andrew Jones, Sudipto Banerjee, Barbara E. Engelhardt
摘要
Gaussian processes are pervasive in functional data analysis, machine learning, and spatial statistics for modeling complex dependencies. Scientific data are often heterogeneous in their inputs and contain multiple known discrete groups of samples; thus, it is desirable to leverage the similarity among groups while accounting for heterogeneity across groups. We propose multi-group Gaussian processes (MGGPs) defined over $\mathbb{R}^p\times \mathscr{C}$, where $\mathscr{C}$ is a finite set representing the group label, by developing general classes of valid (positive definite) covariance functions on such domains. MGGPs are able to accurately recover relationships between the groups and efficiently share strength across samples from all groups during inference, while capturing distinct group-specific behaviors in the conditional posterior distributions. We demonstrate inference in MGGPs through simulation experiments, and we apply our proposed MGGP regression framework to gene expression data to illustrate the behavior and enhanced inferential capabilities of multi-group Gaussian processes by jointly modeling continuous and categorical variables.
Econometrica Bayesian Methods · 发表于 2025
Leonardo Iacovone, David McKenzie, Rachael Meager
摘要
Policymakers often test expensive new programs on relatively small samples. Formally incorporating informative Bayesian priors into impact evaluation offers the promise to learn more from these experiments. We evaluate a Colombian program for 200 firms which aimed to increase exporting. Priors were elicited from academics, policymakers, and firms. Contrary to these priors, frequentist estimation cannot reject null effects in 2019, and finds some negative impacts in 2020. For binary outcomes like whether firms export, frequentist estimates are relatively precise, and Bayesian posterior intervals update to overlap almost completely with standard confidence intervals. For outcomes like increasing export variety, where the priors align with the data, the value of these priors is seen in posterior intervals that are considerably narrower than the confidence intervals. Finally, for noisy outcomes like export value, posterior intervals show almost no updating from priors, highlighting how uninformative the data are about such outcomes. Future policy experiments could use these posteriors as priors in a Bayesian or empirical Bayesian analysis.
JMLR Bayesian Methods · 发表于 2025
Rajarshi Guhaniyogi, Laura Baracaldo, Sudipto Banerjee
摘要
Varying coefficient models are popular for estimating nonlinear regression functions in functional data models. Their Bayesian variants have received limited attention in large data applications, primarily due to prohibitively slow posterior computations using Markov chain Monte Carlo (MCMC) algorithms. We introduce Bayesian data sketching for varying coefficient models to obviate computational challenges presented by large sample sizes. To address the challenges of analyzing large data, we compress the functional response vector and predictor matrix by a random linear transformation to achieve dimension reduction and conduct inference on the compressed data. Our approach distinguishes itself from several existing methods for analyzing large functional data in that it requires neither the development of new models or algorithms nor any specialized computational hardware while delivering fully model-based Bayesian inference. Well-established methods and algorithms for varying-coefficient regression models can be applied to the compressed data. We establish posterior contraction rates for estimating the varying coefficients and predicting the outcome at new locations with the randomly compressed data model. We use simulation experiments and analyze remote sensed vegetation data to empirically illustrate the inferential and computational efficiency of our approach.
JMLR Bayesian Methods · 发表于 2025
Charles Riou, Pierre Alquier, Badr-Eddine Chérief-Abdellatif
摘要
Bernstein's condition is a key assumption that guarantees fast rates in machine learning. For example, under this condition, the Gibbs posterior with prior $\pi$ has an excess risk in $O(d_{\pi}/n)$, as opposed to $O(\sqrt{d_{\pi}/n})$ in the general case, where $n$ denotes the number of observations and $d_{\pi}$ is a complexity parameter which depends on the prior $\pi$. In this paper, we examine the Gibbs posterior in the context of meta-learning, i.e., when learning the prior $\pi$ from $T$ previous tasks. Our main result is that Bernstein's condition always holds at the meta level, regardless of its validity at the observation level. This implies that the additional cost to learn the Gibbs prior $\pi$, which will reduce the term $d_\pi$ across tasks, is in $O(1/T)$, instead of the expected $O(1/\sqrt{T})$. We further illustrate how this result improves on the standard rates in three different settings: discrete priors, Gaussian priors and mixture of Gaussian priors.
JMLR Latent Variables & Mixtures · 发表于 2025
Hanyuan Hang
摘要
In this paper, we propose an ensemble learning algorithm named bagged $k$-distance for mode-based clustering (BDMBC) by putting forward a new measure called the probability of localized level sets (PLLS), which enables us to find all clusters for varying densities with a global threshold. On the theoretical side, we show that with a properly chosen number of nearest neighbors $k_D$ in the bagged $k$-distance, the sub-sample size $s$, the bagging rounds $B$, and the number of nearest neighbors $k_L$ for the localized level sets, BDMBC can achieve optimal convergence rates for mode estimation. It turns out that with a relatively small $B$, the sub-sample size $s$ can be much smaller than the number of training data $n$ at each bagging round, and the number of nearest neighbors $k_D$ can be reduced simultaneously. Moreover, we establish fast convergence rates for the level set estimation of the PLLS in terms of Hausdorff distance, which reveals that BDMBC can find localized level sets for varying densities and thus enjoys local adaptivity. On the practical side, we conduct numerical experiments to empirically verify the effectiveness of BDMBC for mode estimation and level set estimation, which demonstrates the promising accuracy and efficiency of our proposed algorithm.
JMLR Other · 发表于 2025
Yuchao Cai, Hanfang Yang, Yuheng Ma, Hanyuan Hang
摘要
We consider the paradigm of unsupervised anomaly detection, which involves the identification of anomalies within a dataset in the absence of labeled examples. Though distance-based methods are top-performing for unsupervised anomaly detection, they suffer heavily from the sensitivity to the choice of the number of the nearest neighbors. In this paper, we propose a new distance-based algorithm called bagged regularized $k$-distances for anomaly detection (BRDAD), converting the unsupervised anomaly detection problem into a convex optimization problem. Our BRDAD algorithm selects the weights by minimizing the surrogate risk, i.e., the finite sample bound of the empirical risk of the bagged weighted $k$-distances for density estimation (BWDDE). This approach enables us to successfully address the sensitivity challenge of the hyperparameter choice in distance-based algorithms. Moreover, when dealing with large-scale datasets, the efficiency issues can be addressed by the incorporated bagging technique in our BRDAD algorithm. On the theoretical side, we establish fast convergence rates of the AUC regret of our algorithm and demonstrate that the bagging technique significantly reduces the computational complexity. On the practical side, we conduct numerical experiments to illustrate the insensitivity of the parameter selection of our algorithm compared with other state-of-the-art distance-based methods. Furthermore, our method achieves superior performance on real-world datasets with the introduced bagging technique compared to other approaches.
JMLR Other · 发表于 2025
Frank H. van der Meulen, S. Sommer
摘要
We study smoothing for discrete- and continuous-time stochastic processes on directed acyclic graphs (DAGs) when observations are available only at the leaf nodes, a problem common in phylogenetics, epidemiology, and signal processing. We introduce a unified framework built around guiding (also called twisting): a change-of-measure defined by guiding functions that transforms the original process into a guided process whose distribution approximates the smoothing distribution. The Radon-Nikodym derivative quantifies the discrepancy between the two measures. On directed trees, guiding functions are obtained via a backward-filtering step. By isolating backward filtering and forward guiding as elementary operations, we show that the approach extends beyond traditional state-space models and particle filters. We also generalize guiding to edges governed by continuous-time dynamics, using the change-of-measure construction described by Palmowski and Rolski (2002). The versatility of the framework is illustrated with two numerical examples: (i) a diffusion model for shape deformation on a tree, and (ii) inference in a factorial hidden Markov model.
JMLR Other · 发表于 2025
Justin Bunker, Mark Girolami, Hefin Lambley, Andrew M. Stuart, T. J. Sullivan
摘要
Autoencoders have found widespread application in both their original deterministic form and in their variational formulation (VAEs). In scientific applications and in image processing it is often of interest to consider data that are viewed as functions; while discretisation (of differential equations arising in the sciences) or pixellation (of images) renders problems finite dimensional in practice, conceiving first of algorithms that operate on functions, and only then discretising or pixellating, leads to better algorithms that smoothly operate between resolutions. In this paper function-space versions of the autoencoder (FAE) and variational autoencoder (FVAE) are introduced, analysed, and deployed. Well-definedness of the objective governing VAEs is a subtle issue, particularly in function space, limiting applicability. For the FVAE objective to be well defined requires compatibility of the data distribution with the chosen generative model; this can be achieved, for example, when the data arise from a stochastic differential equation, but is generally restrictive. The FAE objective, on the other hand, is well defined in many situations where FVAE fails to be. Pairing the FVAE and FAE objectives with neural operator architectures that can be evaluated on any mesh enables new applications of autoencoders to inpainting, superresolution, and generative modelling of scientific data.
Econometrica Econometrics & Economics · 发表于 2025
Tingjun Liu, Dan Bernhardt
摘要
We consider a classical auction setting in which an asset/project is sold to buyers who privately receive signals about expected payoffs, and payoffs are more sensitive to a bidder's signal if he runs the project than if another bidder does. We show that a seller can increase revenues by sometimes allocating cash‐flow rights and control to different bidders, for example, with the highest bidder receiving cash flows and the second‐highest receiving control. Separation reduces a bidder's information rent, which depends on the importance of his private information for the value of his awarded cash flows. As project payoffs are most sensitive to a bidder's information if he controls the project, allocating cash flow to another bidder lowers bidders' informational advantage. As a result, when signals are close, the seller can increase revenues by splitting rights between the top two bidders.
JMLR High-Dimensional · 发表于 2025
Daiqi Gao, Yufeng Liu, Donglin Zeng
摘要
Dynamic treatment regimes or policies are a sequence of decision functions over multiple stages that are tailored to individual features. One important class of treatment policies in practice, namely multi-stage stationary treatment policies, prescribes treatment assignment probabilities using the same decision function across stages, where the decision is based on the same set of features consisting of time-evolving variables (e.g., routinely collected disease biomarkers). Although there has been extensive literature on constructing valid inference for the value function associated with dynamic treatment policies, little work has focused on the policies themselves, especially in the presence of high-dimensional features. We aim to fill the gap in this work. Specifically, we first obtain the multi-stage stationary treatment policy by minimizing the negative augmented inverse probability weighted estimator of the value function to increase asymptotic efficiency. An $L_1$ penalty is applied on the policy parameters to select important features. We then construct one-step improvements of the policy parameter estimators for valid inference. Theoretically, we show that the improved estimators are asymptotically normal, even if nuisance parameters are estimated at a slow convergence rate and the dimension of the features increases with the sample size. Our numerical studies demonstrate that the proposed method estimates a sparse policy with a near-optimal value function and conducts valid inference for the policy parameters.
JMLR Other · 发表于 2025
Jiacheng Miao, Xinran Miao, Yixuan Wu, Jiwei Zhao, Qiongshi Lu
摘要
A primary challenge facing modern scientific research is the limited availability of gold-standard data, which can be costly, labor-intensive, or invasive to obtain. With the rapid development of machine learning (ML), scientists can now employ ML algorithms to predict gold-standard outcomes using variables that are easier to obtain. However, these predicted outcomes are often used directly in subsequent statistical analyses, ignoring imprecision and heterogeneity introduced by the prediction procedure. This will likely result in false positive findings and invalid scientific conclusions. In this work, we introduce PoSt-Prediction Adaptive inference (PSPA) that allows valid and powerful inference based on ML-predicted data. Its “assumption-lean” property guarantees reliable statistical inference without assumptions on the ML prediction. Its “data-adaptive” feature guarantees an efficiency gain over existing methods, regardless of the accuracy of ML prediction. We demonstrate the statistical superiority and broad applicability of our method through simulations and real-data applications.
JMLR Other · 发表于 2025
Pierre-Alexandre Mattei, Damien Garreau
摘要
Ensemble methods combine the predictions of several base models. We study whether or not including more models always improves their average performance. This question depends on the kind of ensemble considered, as well as the predictive metric chosen. We focus on situations where all members of the ensemble are a priori expected to perform equally well, which is the case of several popular methods such as random forests or deep ensembles. In this setting, we show that ensembles are getting better all the time if, and only if, the considered loss function is convex. More precisely, in that case, the loss of the ensemble is a decreasing function of the number of models. When the loss function is nonconvex, we show a series of results that can be summarised as: ensembles of good models keep getting better, and ensembles of bad models keep getting worse. To this end, we prove a new result on the monotonicity of tail probabilities that may be of independent interest. We illustrate our results on a medical problem (diagnosing melanomas using neural nets) and a “wisdom of crowds” experiment (guessing the ratings of upcoming movies).
JMLR Latent Variables & Mixtures · 发表于 2025
Ery Arias-Castro, Elizabeth Coda
摘要
In this paper, we take an axiomatic approach to defining a population hierarchical clustering for piecewise constant densities, and in a similar manner to Lebesgue integration, extend this definition to more general densities. When the density satisfies some mild conditions, e.g., when it has connected support, is continuous, and vanishes only at infinity, or when the connected components of the density satisfy these conditions, our axiomatic definition results in Hartigan's definition of cluster tree.
JMLR Other · 发表于 2025
Qi Zhang, Yifei Wang, Yisen Wang
摘要
Recently, self-supervised contrastive learning has achieved great success on various tasks. However, its underlying working mechanism is yet unclear. In this paper, we first provide the tightest bounds based on the widely adopted assumption of conditional independence. Further, we relax the conditional independence assumption to a more practical assumption of augmentation overlap and derive the asymptotically closed bounds for the downstream performance. Our proposed augmentation overlap theory hinges on the insight that the support of different intra-class samples will become more overlapped under aggressive data augmentations, thus simply aligning the positive samples (augmented views of the same sample) could make contrastive learning cluster intra-class samples together. Moreover, from the newly derived augmentation overlap perspective, we develop an unsupervised metric for the representation evaluation of contrastive learning, which aligns well with the downstream performance almost without relying on additional modules. Code is available at https://github.com/PKU-ML/GARC.
JMLR Networks & Graphs Bayesian Methods High-Dimensional · 发表于 2025
Tong Xu, Simge Küçükyavuz, Ali Shojaie, Armeen Taeb
摘要
This paper studies the problem of learning Bayesian networks from continuous observational data, generated according to a linear Gaussian structural equation model. We consider an $\ell_0$-penalized maximum likelihood estimator for this problem, which is known to have favorable statistical properties but is computationally challenging to solve, especially for medium-sized Bayesian networks. We propose a new coordinate descent algorithm to approximate this estimator and prove several remarkable properties of our procedure: The algorithm converges to a coordinate-wise minimum, and despite the non-convexity of the loss function, as the sample size tends to infinity, the objective value of the coordinate descent solution converges to the optimal objective value of the $\ell_0$-penalized maximum likelihood estimator. To the best of our knowledge, our proposal is the first coordinate descent procedure endowed with optimality guarantees in the context of learning Bayesian networks. Numerical experiments on synthetic and real data demonstrate that our coordinate descent method can obtain near-optimal solutions while being scalable.
JMLR Optimization & Learning Theory · 发表于 2025
Qihang Lin, Negar Soheili, Runchao Ma, Selvaprabu Nadarajah
摘要
Recent efforts to accelerate first-order methods have focused on convex optimization problems that satisfy a geometric property known as error-bound condition, which covers a broad class of problems, including piece-wise linear programs and strongly convex programs. Parameter-free first-order methods that employ projection-free updates have the potential to broaden the benefit of acceleration. Such a method has been developed for unconstrained convex optimization but is lacking for general constrained convex optimization. We propose a parameter-free level-set method for the latter constrained case based on projection-free subgradient method that exhibits accelerated convergence for problems that satisfy an error-bound condition. Our method maintains a separate copy of the level-set sub-problem for each level parameter value and restarts the computation of these copies based on objective function progress. Applying such a restarting scheme in a level-set context is novel and results in an algorithm that dynamically adapts the precision of each copy. This property is key to extending prior restarting methods based on static precision that have been proposed for unconstrained convex optimization to handle constraints. We report promising numerical performance relative to benchmark methods.
JMLR Machine Learning Networks & Graphs Optimization & Learning Theory · 发表于 2025
Albert Senen--Cerda, Jaron Sanders
摘要
We investigate the convergence and convergence rate of stochastic training algorithms for Neural Networks (NNs) that have been inspired by Dropout (Hinton et al., 2012). With the goal of avoiding overfitting during training in NNs, dropout algorithms consist in practice of multiplying the weight matrices of a NN componentwise by independently drawn random matrices with $\{0, 1\}$-valued entries during each iteration of Stochastic Gradient Descent (SGD). This paper presents a probability theoretical proof that for fully-connected NNs with differentiable, polynomially bounded activation functions, if we project the weights onto a compact set when using a dropout algorithm, then the weights of the NN converge to a unique stationary point of a projected system of Ordinary Differential Equations (ODEs). After this general convergence guarantee, we go on to investigate the convergence rate of dropout. Firstly, we obtain generic sample complexity bounds for finding $\epsilon$-stationary points of smooth nonconvex functions using SGD with dropout that explicitly depend on the dropout probability. Secondly, we obtain an upper bound on the rate of convergence of Gradient Descent (GD) on the limiting ODEs of dropout algorithms for NNs with the shape of an arborescence of arbitrary depth and with linear activation functions. The latter bound shows that for an algorithm such as Dropout or Dropconnect (Wan et al., 2013), the convergence rate can be impaired exponentially by the depth of the arborescence. In contrast, we experimentally observe no such dependence for wide NNs with just a few dropout layers. We also provide a heuristic argument for this observation. Our results suggest that there is a change of scale of the effect of the dropout probability in the convergence rate that depends on the relative size of the width of the NN compared to its depth.
JMLR Other · 发表于 2025
Wanli Qiao, Wolfgang Polonik
摘要
The extraction of filamentary structure from a point cloud is discussed. The filaments are modeled as ridge lines or higher dimensional ridges of an underlying density. We propose two novel algorithms, and provide theoretical guarantees for their convergences, by which we mean that the algorithms can asymptotically recover the full ridge set. We consider the new algorithms as alternatives to the Subspace Constrained Mean Shift (SCMS) algorithm for which no such theoretical guarantees are known.
JMLR Other · 发表于 2025
Sebastian Krämer
摘要
The affine rank minimization problem is a well-known approach to matrix recovery. While there are various surrogates to this NP-hard problem, we prove that the asymptotic minimization of log-det objective functions indeed always reveals the desired, lowest-rank matrices---whereas such may or may not recover a sought-after ground truth. Concerning commonly applied methods such as iteratively reweighted least squares, one thus remains with two difficult to distinguish concerns: how problematic are local minima inherent to the approach truly; and opposingly, how influential instead is the numerical realization. We first show that comparable solution statements do not hold true for Schatten-$p$ functions, including the nuclear norm, and discuss the role of divergent minimizers. Subsequently, we outline corresponding implications for general optimization approaches as well as the more specific IRLS-$0$ algorithm, emphasizing through examples that the transition of the involved smoothing parameter to zero is frequently a more substantial issue than non-convexity. Lastly, we analyze several presented aspects empirically in a series of numerical experiments. In particular, allowing for instance sufficiently many iterations, one may even observe a phase transition for generic recoverability at the absolute theoretical minimum.
JMLR Optimization & Learning Theory Bayesian Methods · 发表于 2025
Shouri Hu, Haowei Wang, Zhongxiang Dai, Bryan Kian Hsiang Low, Szu Hui Ng
摘要
The expected improvement (EI) is one of the most popular acquisition functions for Bayesian optimization (BO) and has demonstrated good empirical performances in many applications for the minimization of simple regret. However, under the evaluation metric of cumulative regret, the performance of EI may not be competitive, and its existing theoretical regret upper bound still has room for improvement. To adapt the EI for better performance under cumulative regret, we introduce a novel quantity called the evaluation cost which is compared against the acquisition function, and with this, develop the expected improvement-cost (EIC) algorithm. In each iteration of EIC, a new point with the largest acquisition function value is sampled, only if that value exceeds its evaluation cost. If none meets this criteria, the current best point is resampled. This evaluation cost quantifies the potential downside of sampling a point, which is important under the cumulative regret metric as the objective function value in every iteration affects the performance measure. We establish in theory a high-probability regret upper bound of EIC based on the maximum information gain, which is tighter than the bound of existing EI-based algorithms. It is also comparable to the regret bound of other popular BO algorithms such as Thompson sampling (GP-TS) and upper confidence bound (GP-UCB). We further perform experiments to illustrate the improvement of EIC over several popular BO algorithms.
Econometrica Econometrics & Economics Optimization & Learning Theory · 发表于 2025
Nicolò Cesa-Bianchi, Roberto Colomboni, Maximilian Kasy
摘要
We consider the problem of repeatedly choosing policies to maximize social welfare. Welfare is a weighted sum of private utility and public revenue. Earlier outcomes inform later policies. Utility is not observed, but indirectly inferred. Response functions are learned through experimentation. We derive a lower bound on regret, and a matching adversarial upper bound for a variant of the Exp3 algorithm. Cumulative regret grows at a rate of T 2/3. This implies that (i) welfare maximization is harder than the multiarmed bandit problem (with a rate of T 1/2 for finite policy sets), and (ii) our algorithm achieves the optimal rate. For the stochastic setting, if social welfare is concave, we can achieve a rate of T 1/2 (for continuous policy sets), using a dyadic search algorithm. We analyze an extension to nonlinear income taxation, and sketch an extension to commodity taxation. We compare our setting to monopoly pricing (which is easier), and price setting for bilateral trade (which is harder).
JMLR Meta-Analysis & Data Integration Nonparametric & Semiparametric Econometrics & Economics · 发表于 2025
Shao-Bo Lin, Xiaotong Liu, Di Wang, Hai Zhang, Ding-Xuan Zhou
摘要
Data silos, mainly caused by privacy and interoperability, significantly constrain collaborations among different organizations with similar data for the same purpose. Distributed learning based on divide-and-conquer provides a promising way to settle the data silos, but it suffers from several challenges, including autonomy, privacy guarantees, and the necessity of collaborations. This paper focuses on developing an adaptive distributed kernel ridge regression (AdaDKRR) by taking autonomy in parameter selection, privacy in communicating non-sensitive information, and the necessity of collaborations for performance improvement into account. We provide both solid theoretical verifications and comprehensive experiments for AdaDKRR to demonstrate its feasibility and effectiveness. Theoretically, we prove that under some mild conditions, AdaDKRR performs similarly to running the optimal learning algorithms on the whole data, verifying the necessity of collaborations and showing that no other distributed learning scheme can essentially beat AdaDKRR under the same conditions. Numerically, we test AdaDKRR on both toy simulations and two real-world applications to show that AdaDKRR is superior to other existing distributed learning schemes. All these results show that AdaDKRR is a feasible scheme to overcome data silos, which are highly desired in numerous application regions such as intelligent decision-making, pricing forecasting, and performance prediction for products.
JMLR Optimization & Learning Theory Meta-Analysis & Data Integration · 发表于 2025
Boxin Zhao, Lingxiao Wang, Ziqi Liu, Zhiqiang Zhang, Jun Zhou, Chaochao Chen, Mladen Kolar
摘要
Due to the high cost of communication, federated learning (FL) systems need to sample a subset of clients that are involved in each round of training. As a result, client sampling plays an important role in FL systems as it affects the convergence rate of optimization algorithms used to train machine learning models. Despite its importance, there is limited work on how to sample clients effectively. In this paper, we cast client sampling as an online learning task with bandit feedback, which we solve with an online stochastic mirror descent (OSMD) algorithm designed to minimize the sampling variance. We then theoretically show how our sampling method can improve the convergence speed of federated optimization algorithms over the widely used uniform sampling. Through both simulated and real data experiments, we empirically illustrate the advantages of the proposed client sampling algorithm over uniform sampling and existing online learning-based sampling strategies. The proposed adaptive sampling procedure is applicable beyond the FL problem studied here and can be used to improve the performance of stochastic optimization procedures such as stochastic gradient descent and stochastic coordinate descent.
Econometrica Other · 发表于 2025
Timothy B. Armstrong, Patrick Kline, Liyang Sun
摘要
Empirical research typically involves a robustness‐efficiency tradeoff. A researcher seeking to estimate a scalar parameter can invoke strong assumptions to motivate a restricted estimator that is precise but may be heavily biased, or they can relax some of these assumptions to motivate a more robust, but variable, unrestricted estimator. When a bound on the bias of the restricted estimator is available, it is optimal to shrink the unrestricted estimator towards the restricted estimator. For settings where a bound on the bias of the restricted estimator is unknown, we propose adaptive estimators that minimize the percentage increase in worst‐case risk relative to an oracle that knows the bound. We show that adaptive estimators solve a weighted convex minimax problem and provide lookup tables facilitating their rapid computation. Revisiting some well‐known empirical studies where questions of model specification arise, we examine the advantages of adapting to—rather than testing for—misspecification.
JMLR Optimization & Learning Theory · 发表于 2025
Noufel FRIKHA, Maximilien GERMAIN, Mathieu LAURIERE, Huyen PHAM, Xuanye SONG
摘要
We study policy gradient for mean-field control in continuous time in a reinforcement learning setting. By considering randomised policies with entropy regularisation, we derive a gradient expectation representation of the value function, which is amenable to actor-critic type algorithms, where the value functions and the policies are learnt alternately based on observation samples of the state and model-free estimation of the population state distribution, either by offline or online learning. In the linear-quadratic mean-field framework, we obtain an exact parametrisation of the actor and critic functions defined on the Wasserstein space. Finally, we illustrate the results of our algorithms with some numerical experiments on concrete examples.
JMLR Optimization & Learning Theory · 发表于 2025
Shi Chen, Qin Li, Oliver Tse, Stephen J. Wright
摘要
The acceleration of gradient-based optimization methods is a subject of significant practical and theoretical importance, particularly within machine learning applications. While much attention has been directed towards optimizing within Euclidean space, the need to optimize over spaces of probability measures in machine learning motivates the exploration of accelerated gradient methods in this context, too. To this end, we introduce a Hamiltonian-flow approach analogous to momentum-based approaches in Euclidean space. We demonstrate that, in the continuous-time setting, algorithms based on this approach can achieve convergence rates of arbitrarily high order. We complement our findings with numerical examples.
JMLR Latent Variables & Mixtures Networks & Graphs Optimization & Learning Theory · 发表于 2025
Yves Atchade, Xinru Liu, Qiuyun Zhu
摘要
We consider inverse problems where the forward model, that is the conditional distribution of the observation ${\bf y}\in\mathbb{R}^{d_y}$ given the latent variable of interest ${\bf x}\in\mathbb{R}^{d_x}$ is known, and access is given to a data set in which multiple instances of $({\bf x},{\bf y})$ are observed. In this context, algorithm unrolling has become a very popular approach for designing state-of-the-art deep neural network architectures that effectively exploit the forward model. We analyze the statistical properties of the gradient descent network (GDN), a well-known architecture driven by proximal gradient descent that epitomizes unrolling learning. Under some regularity conditions, we show that when $d_y\geq d_x$, the GDN estimator solves the inverse problem at a statistical rate faster than the nonparametric minimax rate achievable while ignoring the forward model. Furthermore, when the negative log-density of the latent variable ${\bf x}$ has a simple proximal operator, we show that GDN achieves the parametric rate $O(1/\sqrt{n})$. Furthermore, our results are explicit in the unrolling depth of the network and suggest that unrolling models are typically prone to overfitting as the unrolling depth increases, and careful tuning as function of the sample size is required for best performances. We provide several examples to illustrate these results.
JMLR Machine Learning · 发表于 2025
Samuel E. Otto, Nicholas Zolman, J. Nathan Kutz, Steven L. Brunton
摘要
Symmetry is present throughout nature and continues to play an increasingly central role in machine learning. In this paper, we provide a unifying theoretical and methodological framework for incorporating Lie group symmetry into machine learning models in three ways: 1. enforcing known symmetry when training a model; 2. discovering unknown symmetries of a given model or data set; and 3. promoting symmetry during training by learning a model that breaks symmetries within a user-specified candidate group only when the data provide sufficient evidence. We show that these tasks can be cast within a common mathematical framework whose central object is the Lie derivative. We extend and unify several existing results by showing that enforcing and discovering symmetry are linear-algebraic tasks that are dual under the bilinear pairing induced by the Lie derivative. We also propose a novel way to promote symmetry by introducing a class of convex regularizers, built from the Lie derivative with a nuclear-norm relaxation, that penalizes symmetry breaking during training. We explain how these ideas can be applied to a wide range of machine learning models including basis-function regression, dynamical-systems discovery, neural networks, and neural operators acting on fields.
JMLR Optimization & Learning Theory · 发表于 2025
Lukas Zierahn, Dirk van der Hoeven, Tal Lancewicki, Aviv Rosenberg, Nicolò Cesa-Bianchi
摘要
We derive a new analysis of Follow The Regularized Leader (FTRL) for online learning with delayed bandit feedback. By separating the cost of delayed feedback from that of bandit feedback, our analysis allows us to obtain new results in four important settings. We derive the first optimal (up to logarithmic factors) regret bounds for combinatorial semi-bandits with delay and adversarial Markov Decision Processes with delay (both known and unknown transition functions). Furthermore, we use our analysis to develop an efficient algorithm for linear bandits with delay achieving near-optimal regret bounds. In order to derive these results we show that FTRL remains stable across multiple rounds under mild assumptions on the regularizer.