arXiv subjects

Haixiang Zhang

Publications and source records attributed to Haixiang Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Distributionally Robust Joint Chance-Constrained Optimal Power Flow using Relative Entropy

Designing robust algorithms for the optimal power flow (OPF) problem is critical for the control of large-scale power systems under uncertainty. The chance-constrained OPF (CCOPF) problem provides a natural formulation of the trade-off between the operating cost and the constraint satisfaction rate. In this work, we propose a new data-driven algorithm for the CCOPF problem, based on distributionally robust optimization (DRO). \revise{We show that the proposed reformulation of the distributionally robust chance constraints is exact, whereas other approaches in the CCOPF literature rely on conservative approximations. We establish out-of-sample robustness guarantees for the distributionally robust solution and prove that the solution is the most efficient among all approaches enjoying the same guarantees.} We apply the proposed algorithm to the the CCOPF problem and compare the performance of our approach with existing methods using simulations on IEEE benchmark power systems.

math.OC↗

arXiv2024

An Online Updating Approach for Estimating and Testing Mediation Effects with Big Data Streams

The use of mediation analysis has become increasingly popular in various research fields in recent years. The primary objective of mediation analysis is to examine the indirect effects along the pathways from exposure to outcome. Meanwhile, the advent of data collection technology has sparked a surge of interest in the burgeoning field of big data analysis, where mediation analysis of streaming data sets has recently garnered significant attention. The enormity of the data, however, results in an augmented computational burden. The present study proposes an online updating approach to address this issue, aiming to estimate and test mediation effects in the context of linear and logistic mediation models with massive data streams. The proposed algorithm significantly enhances the computational efficiency of Sobel test, adjusted Sobel test, joint significance test, and adjusted joint significance test. We conduct a substantial number of numerical simulations to evaluate the performance of the renewable method. Two real-world examples are employed to showcase the practical applicability of this approach.

stat.CO↗

arXiv2024

DsubCox: A Fast Subsampling Algorithm for Cox Model with Distributed and Massive Survival Data

To ensure privacy protection and alleviate computational burden, we propose a fast subsmaling procedure for the Cox model with massive survival datasets from multi-centered, decentralized sources. The proposed estimator is computed based on optimal subsampling probabilities that we derived and enables transmission of subsample-based summary level statistics between different storage sites with only one round of communication. For inference, the asymptotic properties of the proposed estimator were rigorously established. An extensive simulation study demonstrated that the proposed approach is effective. The methodology was applied to analyze a large dataset from the U.S. airlines.

stat.CO↗

arXiv2024

Refitted cross-validation estimation for high-dimensional subsamples from low-dimension full data

The technique of subsampling has been extensively employed to address the challenges posed by limited computing resources and meet the needs for expedite data analysis. Various subsampling methods have been developed to meet the challenges characterized by a large sample size with a small number of parameters. However, direct applications of these subsampling methods may not be suitable when the dimension is also high and available computing facilities at hand are only able to analyze a subsample of size similar or even smaller than the dimension. In this case, although there is no high-dimensional problem in the full data, the subsample may have a sample size smaller or smaller than the number of parameters, making it a high-dimensional problem. We call this scenario the high-dimensional subsample from low-dimension full data problem. In this paper, we tackle this problem by proposing a novel subsampling-based approach that combines penalty-based dimension reduction and refitted cross-validation. The asymptotic normality of the refitted cross-validation subsample estimator is established, which plays a crucial role in statistical inference. The proposed method demonstrates appealing performance in numerical experiments on simulated data and a real data application.

stat.ME↗

arXiv2024

Efficient Adjusted Joint Significance Test and Sobel-Type Confidence Interval for Mediation Effect

Mediation analysis is an important statistical tool in many research fields, where the joint significance test is widely utilized for examining mediation effects. Nevertheless, the limitation of this mediation testing method stems from its conservative Type I error, which reduces its statistical power and imposes certain constraints on its utility. The proposed solution to address this gap is the adjusted joint significance test for one mediator, which introduces a novel data-adjusted approach for assessing mediation effects that showcases significant advancements. The method is specifically designed to be user-friendly, thereby eliminating the necessity for intricate procedures. We further extend the adjusted joint significance test for small-scale mediation hypotheses with family-wise error rate (FWER) control. Additionally, a novel adjusted Sobel-type confidence interval is proposed for the mediation effects, demonstrating significant advancements over conventional Sobel's method. The effectiveness of our mediation testing and confidence interval estimation is assessed through extensive simulations, and compared against a multitude of existing approaches. Finally, we present the application of the method to three substantive datasets with continuous, binary and time-to-event outcomes, respectively.

stat.ME↗

arXiv2024

Constraints on Cosmological Models from Quasars Calibrated with Type Ia Supernova by a Gaussian Process

In this paper, we use quasars calibrated from type Ia supernova (SN Ia) to constrain cosmological models. We consider three different X-ray luminosity ($L_{X}$) - ultraviolet luminosity ($L_{UV}$) relations of quasars, i.e., the standard $L_{X}$-$L_{UV}$ relation and two redshift-evolutionary relations (Type I and Type II) respectively constructed from copula and considering a redshift correction to the luminosity of quasars. Only in the case of the Type I relation, quasars can always provide effective constraints on the $Λ$CDM model. Furthermore, we show that, when the observational Hubble data (OHD) are added, the constraints on the absolute magnitude $M$ of SN Ia and the Hubble constant $H_0$ can be obtained. In the $Λ$CDM model, the OHD measurements plus quasars with the Type I relation yields $M$ =$-19.321^{+0.085}_{-0.076}$, which is in good agreement with the measurement from SH0ES ($M=-19.253\pm{0.027}$), and $H_0$ = $70.80\pm3.6~\mathrm{km~s^{-1}Mpc^{-1}}$, falling between the measurements from SH0ES and the Planck cosmic microwave background radiation data.

astro-ph.CO↗

arXiv2024

Exact Recovery for System Identification with More Corrupt Data than Clean Data

This paper investigates the system identification problem for linear discrete-time systems under adversaries and analyzes two lasso-type estimators. We examine both asymptotic and non-asymptotic properties of these estimators in two separate scenarios, corresponding to deterministic and stochastic models for the attack times. Since the samples collected from the system are correlated, the existing results on lasso are not applicable. We prove that when the system is stable and attacks are injected periodically, the sample complexity for exact recovery of the system dynamics is linear in terms of the dimension of the states. When adversarial attacks occur at each time instance with probability p, the required sample complexity for exact recovery scales polynomially in the dimension of the states and the probability p. This result implies almost sure convergence to the true system dynamics under the asymptotic regime. As a by-product, our estimators still learn the system correctly even when more than half of the data is compromised. We highlight that the attack vectors are allowed to be correlated with each other in this work, whereas we make some assumptions about the times at which the attacks happen. This paper provides the first mathematical guarantee in the literature on learning from correlated data for dynamical systems in the case when there is less clean data than corrupt data.

cs.LG↗

arXiv2023

A New Complexity Metric for Nonconvex Rank-one Generalized Matrix Completion

In this work, we develop a new complexity metric for an important class of low-rank matrix optimization problems in both symmetric and asymmetric cases, where the metric aims to quantify the complexity of the nonconvex optimization landscape of each problem and the success of local search methods in solving the problem. The existing literature has focused on two complexity bounds. The RIP constant is commonly used to characterize the complexity of matrix sensing problems. On the other hand, the incoherence and the sampling rate are used when analyzing matrix completion problems. The proposed complexity metric has the potential to generalize these two notions and also applies to a much larger class of problems. To mathematically study the properties of this metric, we focus on the rank-$1$ generalized matrix completion problem and illustrate the usefulness of the new complexity metric on three types of instances, namely, instances with the RIP condition, instances obeying the Bernoulli sampling model, and a synthetic example. We show that the complexity metric exhibits a consistent behavior in the three cases, even when other existing conditions fail to provide theoretical guarantees. These observations provide a strong implication that the new complexity metric has the potential to generalize various conditions of optimization complexity proposed for different applications. Furthermore, we establish theoretical results to provide sufficient and necessary conditions for the existence of spurious solutions in terms of the proposed complexity metric. This contrasts with the RIP and incoherence conditions that fail to provide any necessary condition.

math.OC↗

arXiv2023

Approximating Partial Likelihood Estimators via Optimal Subsampling

With the growing availability of large-scale biomedical data, it is often time-consuming or infeasible to directly perform traditional statistical analysis with relatively limited computing resources at hand. We propose a fast subsampling method to effectively approximate the full data maximum partial likelihood estimator in Cox's model, which largely reduces the computational burden when analyzing massive survival data. We establish consistency and asymptotic normality of a general subsample-based estimator. The optimal subsampling probabilities with explicit expressions are determined via minimizing the trace of the asymptotic variance-covariance matrix for a linearly transformed parameter estimator. We propose a two-step subsampling algorithm for practical implementation, which has a significant reduction in computing time compared to the full data method. The asymptotic properties of the resulting two-step subsample-based estimator is also established. Extensive numerical experiments and a real-world example are provided to assess our subsampling strategy.

stat.ME↗

arXiv2023

A Framework for Mediation Analysis with Massive Data

During the past few years, mediation analysis has gained increasing popularity across various research fields. The primary objective of mediation analysis is to examine the direct impact of exposure on outcome, as well as the indirect effects that occur along the pathways from exposure to outcome. There has been a great number of articles that applied mediation analysis to data from hundreds or thousands of individuals. With the rapid development of technology, the volume of avaliable data increases exponentially, which brings new challenges to researchers. Directly conducting statistical analysis for large datasets is often computationally infeasible. Nonetheless, there is a paucity of findings regarding mediation analysis in the context of big data. In this paper, we propose utilizing subsampled double bootstrap and divide-and-conquer algorithms to conduct statistical mediation analysis on large-scale datasets. The proposed algorithms offer a significant enhancement in computational efficiency over traditional bootstrap confidence interval and Sobel test, while simultaneously ensuring desirable confidence interval coverage and power. We conducted extensive numerical simulations to evaluate the performance of our method. The practical applicability of our approach is demonstrated through two real-world data examples.

stat.CO↗

arXiv2022

Renewable Learning for Multiplicative Regression with Streaming Datasets

When large amounts of data continuously arrive in streams, online updating is an effective way to reduce storage and computational burden. The key idea of online updating is that the previous estimators are sequentially updated only using the current data and some summary statistics of historical raw data. In this article, we develop a renewable learning method for a multiplicative regression model with streaming data, where the parameter estimator based on a least product relative error criterion is renewed without revisiting any historical raw data. Under some regularity conditions, we establish the consistency and asymptotic normality of the renewable estimator. Moreover, the theoretical results confirm that the proposed renewable estimator achieves the same asymptotic distribution as the least product relative error estimator with the entire dataset. Numerical studies and two real data examples are provided to evaluate the performance of our proposed method.

stat.ME↗

arXiv2022

A Latent Logistic Regression Model with Graph Data

Recently, graph (network) data is an emerging research area in artificial intelligence, machine learning and statistics. In this work, we are interested in whether node's labels (people's responses) are affected by their neighbor's features (friends' characteristics). We propose a novel latent logistic regression model to describe the network dependence with binary responses. The key advantage of our proposed model is that a latent binary indicator is introduced to indicate whether a node is susceptible to the influence of its neighbour. A score-type test is proposed to diagnose the existence of network dependence. In addition, an EM-type algorithm is used to estimate the model parameters under network dependence. Extensive simulations are conducted to evaluate the performance of our method. Two public datasets are used to illustrate the effectiveness of the proposed latent logistic regression model.

stat.ME↗

arXiv2022

Local and Global Linear Convergence of General Low-rank Matrix Recovery Problems

We study the convergence rate of gradient-based local search methods for solving low-rank matrix recovery problems with general objectives in both symmetric and asymmetric cases, under the assumption of the restricted isometry property. First, we develop a new technique to verify the Polyak-Lojasiewicz inequality in a neighborhood of the global minimizers, which leads to a local linear convergence region for the gradient descent method. Second, based on the local convergence result and a sharp strict saddle property proven in this paper, we present two new conditions that guarantee the global linear convergence of the perturbed gradient descent method. The developed local and global convergence results provide much stronger theoretical guarantees than the existing results. As a by-product, this work significantly improves the existing bounds on the RIP constant required to guarantee the non-existence of spurious solutions.

math.OC↗

arXiv2022

Gradient-based Algorithms for Convex Discrete Optimization via Simulation

We propose new sequential simulation-optimization algorithms for general convex optimization via simulation problems with high-dimensional discrete decision space. The performance of each choice of discrete decision variables is evaluated via stochastic simulation replications. If an upper bound on the overall level of uncertainties is known, our proposed simulation-optimization algorithms utilize the discrete convex structure and are guaranteed with high probability to find a solution that is close to the best within any given user-specified precision level. The proposed algorithms work for any general convex problem and the efficiency is demonstrated by proven upper bounds on simulation costs. The upper bounds demonstrate a polynomial dependence on the dimension and scale of the decision space. For some discrete optimization via simulation problems, a gradient estimator may be available at low costs along with a single simulation replication. By integrating gradient estimators, which are possibly biased, we propose simulation-optimization algorithms to achieve optimality guarantees with a reduced dependence on the dimension under moderate assumptions on the bias.

math.OC↗

arXiv2022

Stochastic Localization Methods for Convex Discrete Optimization via Simulation

We develop and analyze a set of new sequential simulation-optimization algorithms for large-scale multi-dimensional discrete optimization via simulation problems with a convexity structure. The "large-scale" notion refers to that the decision variable has a large number of values to choose from on each dimension. The proposed algorithms are targeted to identify a solution that is close to the optimal solution given any precision level with any given probability. To achieve this target, utilizing the convexity structure, our algorithm design does not need to scan all the choices of the decision variable, but instead sequentially draws a subset of choices of the decision variable and uses them to "localize" potentially near-optimal solutions to an adaptively shrinking region. To show the power of the localization operation, we first consider one-dimensional large-scale problems. We propose the shrinking uniform sampling algorithm, which is proved to achieve the target with an optimal expected simulation cost under an asymptotic criterion. For multi-dimensional problems, we combine the idea of localization with subgradient information and propose a framework to design stochastic cutting-plane methods and the dimension reduction algorithm, whose expected simulation cost have a low dependence on the scale and the dimension of the problems. The proposed algorithms do not require prior information about the Lipschitz constant of the objective function and the simulation costs are upper bounded by a value that is independent of the Lipschitz constant. Finally, we propose an adaptive algorithm to deal with the unknown noise variance case under the assumption that the randomness of the system is Gaussian. We implement the proposed algorithms on both synthetic and queueing simulation optimization problems, and demonstrate better performances compared to benchmark methods.

math.OC↗

arXiv2021

A Mathematical Foundation for Robust Machine Learning based on Bias-Variance Trade-off

A common assumption in machine learning is that samples are independently and identically distributed (i.i.d). However, the contributions of different samples are not identical in training. Some samples are difficult to learn and some samples are noisy. The unequal contributions of samples has a considerable effect on training performances. Studies focusing on unequal sample contributions (e.g., easy, hard, noisy) in learning usually refer to these contributions as robust machine learning (RML). Weighing and regularization are two common techniques in RML. Numerous learning algorithms have been proposed but the strategies for dealing with easy/hard/noisy samples differ or even contradict with different learning algorithms. For example, some strategies take the hard samples first, whereas some strategies take easy first. Conducting a clear comparison for existing RML algorithms in dealing with different samples is difficult due to lack of a unified theoretical framework for RML. This study attempts to construct a mathematical foundation for RML based on the bias-variance trade-off theory. A series of definitions and properties are presented and proved. Several classical learning algorithms are also explained and compared. Improvements of existing methods are obtained based on the comparison. A unified method that combines two classical learning strategies is proposed.

cs.LG↗

arXiv2021

Factorization Approach for Low-complexity Matrix Completion Problems: Exponential Number of Spurious Solutions and Failure of Gradient Methods

It is well-known that the Burer-Monteiro (B-M) factorization approach can efficiently solve low-rank matrix optimization problems under the RIP condition. It is natural to ask whether B-M factorization-based methods can succeed on any low-rank matrix optimization problems with a low information-theoretic complexity, i.e., polynomial-time solvable problems that have a unique solution. In this work, we provide a negative answer to the above question. We investigate the landscape of B-M factorized polynomial-time solvable matrix completion (MC) problems, which are the most popular subclass of low-rank matrix optimization problems without the RIP condition. We construct an instance of polynomial-time solvable MC problems with exponentially many spurious local minima, which leads to the failure of most gradient-based methods. Based on those results, we define a new complexity metric that potentially measures the solvability of low-rank matrix optimization problems based on the B-M factorization approach. In addition, we show that more measurements of the ground truth matrix can deteriorate the landscape, which further reveals the unfavorable behavior of the B-M factorization on general low-rank matrix optimization problems.

math.OC↗

arXiv2021

General Low-rank Matrix Optimization: Geometric Analysis and Sharper Bounds

This paper considers the global geometry of general low-rank minimization problems via the Burer-Monterio factorization approach. For the rank-$1$ case, we prove that there is no spurious second-order critical point for both symmetric and asymmetric problems if the rank-$2$ RIP constant $δ$ is less than $1/2$. Combining with a counterexample with $δ=1/2$, we show that the derived bound is the sharpest possible. For the arbitrary rank-$r$ case, the same property is established when the rank-$2r$ RIP constant $δ$ is at most $1/3$. We design a counterexample to show that the non-existence of spurious second-order critical points may not hold if $δ$ is at least $1/2$. In addition, for any problem with $δ$ between $1/3$ and $1/2$, we prove that all second-order critical points have a positive correlation to the ground truth. Finally, the strict saddle property, which can lead to the polynomial-time global convergence of various algorithms, is established for both the symmetric and asymmetric problems when the rank-$2r$ RIP constant $δ$ is less than $1/3$. The results of this paper significantly extend several existing bounds in the literature.

math.OC↗