Searcharxiv⌕ Search

arXiv subjects

Hongjian Shi

Publications and source records attributed to Hongjian Shi.

15 recordsLinked to original sources
arXiv2026

A Note on the Finite-Exceedance Dependence Principle for Sum--Maximum Asymptotic Independence

Asymptotic independence between sum-type and maximum-type statistics arises repeatedly across many areas of probability and statistics, particularly in high-dimensional testing and random matrix theory. Motivated by the shared structure underlying existing proofs, which is obscured by their model-specific formulations, this note isolates a common intermediate principle. We introduce a finite-exceedance dependence principle: if the finite-exceedance dependence error vanishes at every fixed order and high-order exceedances are mildly controlled, then a finite-order Bonferroni argument yields asymptotic independence between the sum and the maximum. We further give an independent-proxy criterion for verifying this: for each finite set of exceedances, it suffices to construct an exactly independent proxy for the sum-type statistic whose approximation error is negligible in aggregate. Local deletion under dependency graphs and Gaussian regression and residualization are two examples. The framework gives modular derivations of existing sum--max independence results.

math.ST↗
arXiv2026

Combinatorial Arbitrage in Prediction Markets

We develop a framework to identify and quantify arbitrage across logically related prediction-market contracts. We prove that prices outside the convex hull of admissible payoff vectors admit arbitrage and identify optimal arbitrage portfolios through integer programming. Using venue metadata and semantic analysis of resolution criteria, we recover 1.9 million logical relations and compile them into constraint systems covering more than 255,000 contracts. We replay full-depth Polymarket order books to simulate an arbitrageur executing these portfolios against resting orders. Among settled portfolios, an ideal arbitrageur that takes the full displayed depth of every opportunity and reuses its capital at resolution earns a risk-free return of 3.5% a week after spreads and fees. We find that arbitrage opportunities persist longer and attract less observed participation where relation discovery and execution are more costly.

q-fin.TR↗
arXiv2026

AutoSFT: An Agentic System for Automated Data Research in Supervised Fine-Tuning

Language model agents increasingly automate post-training, but when given the full training stack they converge on supervised fine-tuning and spend most effort revising training data. Existing systems give the agent one kind of control, selecting rows from a fixed pool or calling a generator, whereas a human data engineer adjusts proportions, prunes, repairs, replaces and synthesizes the data. We design a data action space for these changes and ask whether an automated system can make the full range rather than one of them, and what structure that requires. We first run this loop with a single agent, and it reaches only one action, adjusting category proportions, while its context fills with execution output and is discarded mid-search. We therefore introduce \textbf{AutoSFT}, a multi-agent system where a meta agent separates execution from analysis, gathers proposals from independent angles, and calls a synthesis subagent for capabilities the pool cannot cover, with the model, the trainer, and the budget fixed within each comparison. Adding each component in turn helps: the executor halves context loss, the analysis subagents lift the best IFEval by $+6.5$, and synthesis adds $+3.4$ IFEval and $+2.9$ IFBench. The same loop transfers to code, where analysis adds $+3.7$ MBPP+ over a single agent that does not improve. Every result comes from runs that retrain from the base model each round, so each component's effect is individually attributable. Our findings suggest that automating SFT data research depends more on covering diverse data actions than on a stronger single action.

cs.AI↗
arXiv2025

Distributed Bilevel Optimization with Dual Pruning for Resource-limited Clients

With the development of large-scale models, traditional distributed bilevel optimization algorithms cannot be applied directly in low-resource clients. The key reason lies in the excessive computation involved in optimizing both the lower- and upper-level functions. Thus, we present the first resource-adaptive distributed bilevel optimization framework with a second-order free hypergradient estimator, which allows each client to optimize the submodels adapted to the available resources. Due to the coupled influence of partial outer parameters x and inner parameters y, it's challenging to theoretically analyze the upper bound regarding the globally averaged hypergradient for full model parameters. The error bound of inner parameter also needs to be reformulated since the local partial training. The provable theorems show that both RABO and RAFBO can achieve an asymptotically optimal convergence rate of $O(1/\sqrt{C_x^{\ast}Q})$, which is dominated by the minimum coverage of the outer parameter $C_x^{\ast}$. Extensive experiments on two different tasks demonstrate the effectiveness and computation efficiency of our proposed methods.

cs.DC↗
arXiv2025

On universal inference in Gaussian mixture models

A recent line of work provides new statistical tools based on game-theory and achieves safe anytime-valid inference without assuming regularity conditions. In particular, the framework of universal inference proposed by Wasserman, Ramdas and Balakrishnan [78] offers new solutions to testing problems by modifying the likelihood ratio test in a data-splitting scheme. In this paper, we study the performance of the resulting split likelihood ratio test under Gaussian mixture models, which are canonical examples for models in which classical regularity conditions fail to hold. We establish that under the null hypothesis, the split likelihood ratio statistic is asymptotically normal with increasing mean and variance. Contradicting the usual belief that the flexibility of universal inference comes at the price of a significant loss of power, we prove that universal inference surprisingly achieves the same detection rate $(n^{-1}\log\log n)^{1/2}$ as the classical likelihood ratio test.

math.ST↗
arXiv2025

Existence of Direct Density Ratio Estimators

Many two-sample problems call for a comparison of two distributions from an exponential family. Density ratio estimation methods provide ways to solve such problems through direct estimation of the differences in natural parameters. The term direct indicates that one avoids estimating both marginal distributions. In this context, we consider the Kullback--Leibler Importance Estimation Procedure (KLIEP), which has been the subject of recent work on differential networks. Our main result shows that the existence of the KLIEP estimator is characterized by whether the average sufficient statistic for one sample belongs to the convex hull of the set of all sufficient statistics for data points in the second sample. For high-dimensional problems it is customary to regularize the KLIEP loss by adding the product of a tuning parameter and a norm of the vector of parameter differences. We show that the existence of the regularized KLIEP estimator requires the tuning parameter to be no less than the dual norm-based distance between the average sufficient statistic and the convex hull. The implications of these existence issues are explored in applications to differential network analysis.

math.ST↗
arXiv2024

Distribution-free tests of multivariate independence based on center-outward quadrant, Spearman, Kendall, and van der Waerden statistics

Due to the lack of a canonical ordering in ${\mathbb R}^d$ for $d>1$, defining multivariate generalizations of the classical univariate ranks has been a long-standing open problem in statistics. Optimal transport has been shown to offer a solution in which multivariate ranks are obtained by transporting data points to a grid that approximates a uniform reference measure (Chernozhukov et al., 2017; Hallin, 2017; Hallin et al., 2021), thereby inducing ranks, signs, and a data-driven ordering of ${\mathbb R}^d$. We take up this new perspective to define and study multivariate analogues of the sign covariance/quadrant statistic, Spearman's rho, Kendall's tau, and van der Waerden covariances. The resulting tests of multivariate independence are fully distribution-free, hence uniformly valid irrespective of the actual (absolutely continuous) distribution of the observations. Our results provide the asymptotic distribution theory for these new test statistics, with asymptotic approximations to critical values to be used for testing independence between random vectors, as well as a power analysis of the resulting tests in an extension of the so-called (bivariate) Konijn model. This power analysis includes a multivariate Chernoff--Savage property guaranteeing that, under elliptical generalized Konijn models, the asymptotic relative efficiency of our van der Waerden tests with respect to Wilks' classical (pseudo-)Gaussian procedure is strictly larger than or equal to one, where equality is achieved under Gaussian distributions only. We similarly provide a lower bound for the asymptotic relative efficiency of our Spearman procedure with respect to Wilks' test, thus extending the classical result by Hodges and Lehmann on the asymptotic relative efficiency, in univariate location models, of Wilcoxon tests with respect to the Student ones.

math.ST↗
arXiv2022

On Azadkia-Chatterjee's conditional dependence coefficient

In recent work, Azadkia and Chatterjee (2021) laid out an ingenious approach to defining consistent measures of conditional dependence. Their fully nonparametric approach forms statistics based on ranks and nearest neighbor graphs. The appealing nonparametric consistency of the resulting conditional dependence measure and the associated empirical conditional dependence coefficient has quickly prompted follow-up work that seeks to study its statistical efficiency. In this paper, we take up the framework of conditional randomization tests (CRT) for conditional independence and conduct a power analysis that considers two types of local alternatives, namely, parametric quadratic mean differentiable alternatives and nonparametric Hölder smooth alternatives. Our local power analysis shows that conditional independence tests using the Azadkia--Chatterjee coefficient remain inefficient even when aided with the CRT framework, and serves as motivation to develop variants of the approach; cf. Lin and Han (2022b). As a byproduct, we resolve a conjecture of Azadkia and Chatterjee by proving central limit theorems for the considered conditional dependence coefficients, with explicit formulas for the asymptotic variances.

math.ST↗
arXiv2021

On universally consistent and fully distribution-free rank tests of vector independence

Rank correlations have found many innovative applications in the last decade. In particular, suitable rank correlations have been used for consistent tests of independence between pairs of random variables. Using ranks is especially appealing for continuous data as tests become distribution-free. However, the traditional concept of ranks relies on ordering data and is, thus, tied to univariate observations. As a result, it has long remained unclear how one may construct distribution-free yet consistent tests of independence between random vectors. This is the problem addressed in this paper, in which we lay out a general framework for designing dependence measures that give tests of multivariate independence that are not only consistent and distribution-free but which we also prove to be statistically efficient. Our framework leverages the recently introduced concept of center-outward ranks and signs, a multivariate generalization of traditional ranks, and adopts a common standard form for dependence measures that encompasses many popular examples. In a unified study, we derive a general asymptotic representation of center-outward rank-based test statistics under independence, extending to the multivariate setting the classical Hájek asymptotic representation results. This representation permits direct calculation of limiting null distributions and facilitates a local power analysis that provides strong support for the center-outward approach by establishing, for the first time, the nontrivial power of center-outward rank-based tests over root-$n$ neighborhoods within the class of quadratic mean differentiable alternatives.

math.ST↗
arXiv2021

On the power of Chatterjee rank correlation

Chatterjee (2021) introduced a simple new rank correlation coefficient that has attracted much recent attention. The coefficient has the unusual appeal that it not only estimates a population quantity first proposed by Dette et al. (2013) that is zero if and only if the underlying pair of random variables is independent, but also is asymptotically normal under independence. This paper compares Chatterjee's new correlation coefficient to three established rank correlations that also facilitate consistent tests of independence, namely, Hoeffding's $D$, Blum-Kiefer-Rosenblatt's $R$, and Bergsma-Dassios-Yanagimoto's $τ^*$. We contrast their computational efficiency in light of recent advances, and investigate their power against local rotation and mixture alternatives. Our main results show that Chatterjee's coefficient is unfortunately rate sub-optimal compared to $D$, $R$, and $τ^*$. The situation is more subtle for a related earlier estimator of Dette et al. (2013). These results favor $D$, $R$, and $τ^*$ over Chatterjee's new correlation coefficient for the purpose of testing independence.

math.ST↗
arXiv2020

Distribution-free consistent independence tests via center-outward ranks and signs

This paper investigates the problem of testing independence of two random vectors of general dimensions. For this, we give for the first time a distribution-free consistent test. Our approach combines distance covariance with the center-outward ranks and signs developed in Hallin (2017). In technical terms, the proposed test is consistent and distribution-free in the family of multivariate distributions with nonvanishing (Lebesgue) probability densities. Exploiting the (degenerate) U-statistic structure of the distance covariance and the combinatorial nature of Hallin's center-outward ranks and signs, we are able to derive the limiting null distribution of our test statistic. The resulting asymptotic approximation is accurate already for moderate sample sizes and makes the test implementable without requiring permutation. The limiting distribution is derived via a more general result that gives a new type of combinatorial non-central limit theorem for double- and multiple-indexed permutation statistics.

math.ST↗
arXiv2020

High dimensional consistent independence testing with maxima of rank correlations

Testing mutual independence for high-dimensional observations is a fundamental statistical challenge. Popular tests based on linear and simple rank correlations are known to be incapable of detecting non-linear, non-monotone relationships, calling for methods that can account for such dependences. To address this challenge, we propose a family of tests that are constructed using maxima of pairwise rank correlations that permit consistent assessment of pairwise independence. Built upon a newly developed Cramér-type moderate deviation theorem for degenerate U-statistics, our results cover a variety of rank correlations including Hoeffding's $D$, Blum-Kiefer-Rosenblatt's $R$, and Bergsma-Dassios-Yanagimoto's $τ^*$. The proposed tests are distribution-free in the class of multivariate distributions with continuous margins, implementable without the need for permutation, and are shown to be rate-optimal against sparse alternatives under the Gaussian copula model. As a by-product of the study, we reveal an identity between the aforementioned three rank correlation statistics, and hence make a step towards proving a conjecture of Bergsma and Dassios.

math.ST↗
arXiv2018

Prostate Segmentation using 2D Bridged U-net

In this paper, we focus on three problems in deep learning based medical image segmentation. Firstly, U-net, as a popular model for medical image segmentation, is difficult to train when convolutional layers increase even though a deeper network usually has a better generalization ability because of more learnable parameters. Secondly, the exponential ReLU (ELU), as an alternative of ReLU, is not much different from ReLU when the network of interest gets deep. Thirdly, the Dice loss, as one of the pervasive loss functions for medical image segmentation, is not effective when the prediction is close to ground truth and will cause oscillation during training. To address the aforementioned three problems, we propose and validate a deeper network that can fit medical image datasets that are usually small in the sample size. Meanwhile, we propose a new loss function to accelerate the learning process and a combination of different activation functions to improve the network performance. Our experimental results suggest that our network is comparable or superior to state-of-the-art methods.

cs.CV↗
arXiv2018

Head Mounted Pupil Tracking Using Convolutional Neural Network

Pupil tracking is an important branch of object tracking which require high precision. We investigate head mounted pupil tracking which is often more convenient and precise than remote pupil tracking, but also more challenging. When pupil tracking suffers from noise like bad illumination, detection precision dramatically decreases. Due to the appearance of head mounted recording device and public benchmark image datasets, head mounted tracking algorithms have become easier to design and evaluate. In this paper, we propose a robust head mounted pupil detection algorithm which uses a Convolutional Neural Network (CNN) to combine different features of pupil. Here we consider three features of pupil. Firstly, we use three pupil feature-based algorithms to find pupil center independently. Secondly, we use a CNN to evaluate the quality of each result. Finally, we select the best result as output. The experimental results show that our proposed algorithm performs better than the present state-of-art.

cs.CV↗