arXiv subjects

Junyong Park

Publications and source records attributed to Junyong Park.

At least 19 recordsLinked to original sources

arXiv2026

Enhancing Differentially Private Mechanisms via Empirical Bayes

Differential privacy (DP) has become the gold standard for ensuring the privacy protection of machine learning and statistical algorithms in recent decades. A plethora of algorithms and methods have been developed to enhance the utility of DP algorithms while maintaining the same level of DP. However, these are often overly complex or computationally ineffective. We propose a novel approach focusing on denoising the output of the simple additive Gaussian mechanism by adopting the idea of \textit{empirical Bayes estimation}. We highlight that the empirical Bayes approach can reduce the mean-squared error solely by taking the output of the Gaussian mechanism as input. Our numerical studies show that this simple yet powerful approach can be applied to improve upon various statistical problems, including histogram release, principal component analysis, and linear regression, often outperforming existing private algorithms.

cs.LG↗

arXiv2026

Online FDR Controlling procedures for statistical SIS Model and its application to COVID19 data

We propose an online false discovery rate (FDR) controlling method based on conditional local FDR (LIS), designed for infectious disease datasets that are discrete and exhibit complex dependencies. Unlike existing online FDR methods, which often assume independence or suffer from low statistical power in dependent settings, our approach effectively controls FDR while maintaining high detection power in realistic epidemic scenarios. For disease modeling, we establish a Dynamic Bayesian Network (DBN) structure within the Susceptible-Infected-Susceptible (SIS) model, a widely used epidemiological framework for infectious diseases. Our method requires no additional tuning parameters apart from the width of the sliding window, making it practical for real-time disease monitoring. From a statistical perspective, we prove that our method ensures valid FDR control under stationary and ergodic dependencies, extending online hypothesis testing to a broader range of dependent and discrete datasets. Additionally, our method achieves higher statistical power than existing approaches by leveraging LIS, which has been shown to be more powerful than traditional $p$-value-based methods. We validate our method through extensive simulations and real-world applications, including the analysis of infectious disease incidence data. Our results demonstrate that the proposed approach outperforms existing methods by achieving higher detection power while maintaining rigorous FDR control.

stat.ME↗

arXiv2026

Box Thirding: Anytime Best Arm Identification under Insufficient Sampling

We introduce Box Thirding (B3), a flexible and efficient algorithm for Best Arm Identification (BAI) under fixed-budget constraints. It is designed for both anytime BAI and scenarios with large N, where the number of arms is too large for exhaustive evaluation within a limited budget T. The algorithm employs an iterative ternary comparison: in each iteration, three arms are compared--the best-performing arm is explored further, the median is deferred for future comparisons, and the weakest is discarded. Even without prior knowledge of T, B3 achieves an epsilon-best arm misidentification probability comparable to Successive Halving (SH), which requires T as a predefined parameter, applied to a randomly selected subset of c0 arms that fit within the budget. Empirical results show that B3 outperforms existing methods under limited-budget constraints in terms of simple regret, as demonstrated on the New Yorker Cartoon Caption Contest dataset.

stat.ML↗

arXiv2026

Two-Stage Multiple Test Procedures Controlling False Discovery Rate with auxiliary variable and their Application to Set4Delta Mutant Data

In this paper, we present novel methodologies that incorporate auxiliary variables for multiple hypotheses testing related to the main point of interest while effectively controlling the false discovery rate. When dealing with multiple tests concerning the primary variable of interest, researchers can use auxiliary variables to set preconditions for the significance of primary variables, thereby enhancing test efficacy. Depending on the auxiliary variable's role, we propose two approaches: one terminates testing of the primary variable if it does not meet predefined conditions, and the other adjusts the evaluation criteria based on the auxiliary variable. Employing the copula method, we elucidate the dependence between the auxiliary and primary variables by deriving their joint distribution from individual marginal distributions.Our numerical studies, compared with existing methods, demonstrate that the proposed methodologies effectively control the FDR and yield greater statistical power than previous approaches solely based on the primary variable. As an illustrative example, we apply our methods to the Set4$Δ$ mutant dataset. Our findings highlight the distinctions between our methodologies and traditional approaches, emphasising the potential advantages of our methods in introducing the auxiliary variable for selecting more genes.

stat.ME↗

arXiv2026

GRTX: Efficient Ray Tracing for 3D Gaussian-Based Rendering

3D Gaussian Splatting has gained widespread adoption across diverse applications due to its exceptional rendering performance and visual quality. While most existing methods rely on rasterization to render Gaussians, recent research has started investigating ray tracing approaches to overcome the fundamental limitations inherent in rasterization. However, current Gaussian ray tracing methods suffer from inefficiencies such as bloated acceleration structures and redundant node traversals, which greatly degrade ray tracing performance. In this work, we present GRTX, a set of software and hardware optimizations that enable efficient ray tracing for 3D Gaussian-based rendering. First, we introduce a novel approach for constructing streamlined acceleration structures for Gaussian primitives. Our key insight is that anisotropic Gaussians can be treated as unit spheres through ray space transformations, which substantially reduces BVH size and traversal overhead. Second, we propose dedicated hardware support for traversal checkpointing within ray tracing units. This eliminates redundant node visits during multi-round tracing by resuming traversal from checkpointed nodes rather than restarting from the root node in each subsequent round. Our evaluation shows that GRTX significantly improves ray tracing performance compared to the baseline ray tracing method with a negligible hardware cost.

cs.GR↗

arXiv2025

FDR Control via Neural Networks under Covariate-Dependent Symmetric Nulls

In modern multiple hypothesis testing, the availability of covariate information alongside the primary test statistics has motivated the development of more powerful and adaptive inference methods. However, most existing approaches rely on p-values that are precomputed under the assumption that their null distributions are independent of the covariates. In this paper, we propose a framework that derives covariate-adaptive p-values from the assumption of a symmetric null distribution of the primary variable given the covariates, without imposing any parametric assumptions. Building on these data-driven p-values, we employ a neural network model to learn a covariate-adaptive rejection threshold via the mirror estimation principle, optimizing the number of discoveries while maintaining valid false discovery rate control. Furthermore, our estimation of the conditional null distribution enables the computation of p-values directly from the raw data. The proposed method provides a principled way to derive covariate-adjusted p-values from raw data and allows seamless integration with previously established p-value based procedures. Simulation studies show that the proposed method outperforms existing approaches in terms of power. We further illustrate its applicability through two real data analyses: age-specific blood pressure data and U.S. air pollution data.

stat.ME↗

arXiv2025

MX+: Pushing the Limits of Microscaling Formats for Efficient Large Language Model Serving

Reduced-precision data formats are crucial for cost-effective serving of large language models (LLMs). While numerous reduced-precision formats have been introduced thus far, they often require intrusive modifications to the software frameworks or are rather unconventional for widespread adoption across hardware vendors. In this paper, we instead focus on recent industry-driven variants of block floating-point (BFP) formats and conduct a comprehensive analysis to push their limits for efficient LLM serving. Our analysis shows that existing ultra low-bit BFP variants struggle to provide reasonable language model performance due to outlier values in blocks. To address the outliers with BFPs, we propose MX+, a cost-effective and non-intrusive extension designed for seamless integration into the microscaling (MX) formats. MX+ builds on the key insight that the outlier does not need to use its exponent field in the element data type, which allows us to repurpose the exponent field as an extended mantissa to increase the precision of the outlier element. Our evaluation shows that MX+ achieves significantly higher model performance compared to the 4-bit MX format (MXFP4) with negligible storage overhead and slowdown, thus offering a compelling alternative to MXFP4 or MXFP6 for efficient LLM inference.

cs.LG↗

arXiv2025

SlotFM: A Motion Foundation Model with Slot Attention for Diverse Downstream Tasks

Wearable accelerometers are used for a wide range of applications, such as gesture recognition, gait analysis, and sports monitoring. Yet most existing foundation models focus primarily on classifying common daily activities such as locomotion and exercise, limiting their applicability to the broader range of tasks that rely on other signal characteristics. We present SlotFM, an accelerometer foundation model that generalizes across diverse downstream tasks. SlotFM uses Time-Frequency Slot Attention, an extension of Slot Attention that processes both time and frequency representations of the raw signals. It generates multiple small embeddings (slots), each capturing different signal components, enabling task-specific heads to focus on the most relevant parts of the data. We also introduce two loss regularizers that capture local structure and frequency patterns, which improve reconstruction of fine-grained details and helps the embeddings preserve task-relevant information. We evaluate SlotFM on 16 classification and regression downstream tasks that extend beyond standard human activity recognition. It outperforms existing self-supervised approaches on 13 of these tasks and achieves comparable results to the best performing approaches on the remaining tasks. On average, our method yields a 4.5% performance gain, demonstrating strong generalization for sensing foundation models.

cs.LG↗

arXiv2025

Silent Impact: Tracking Tennis Shots from the Passive Arm

Wearable technology has transformed sports analytics, offering new dimensions in enhancing player experience. Yet, many solutions involve cumbersome setups that inhibit natural motion. In tennis, existing products require sensors on the racket or dominant arm, causing distractions and discomfort. We propose Silent Impact, a novel and user-friendly system that analyzes tennis shots using a sensor placed on the passive arm. Collecting Inertial Measurement Unit sensor data from 20 recreational tennis players, we developed neural networks that exclusively utilize passive arm data to detect and classify six shots, achieving a classification accuracy of 88.2% and a detection F1 score of 86.0%, comparable to the dominant arm. These models were then incorporated into an end-to-end prototype, which records passive arm motion through a smartwatch and displays a summary of shots on a mobile app. User study (N=10) showed that participants felt less burdened physically and mentally using Silent Impact on the passive arm. Overall, our research establishes the passive arm as an effective, comfortable alternative for tennis shot analysis, advancing user-friendly sports analytics.

cs.HC↗

arXiv2025

FDR controlling procedures with dimension reduction and their application to GWAS with linkage disequilibrium score

Genome-wide association studies (GWAS) have led to the discovery of numerous single nucleotide polymorphisms (SNPs) associated with various phenotypes and complex diseases. However, the identified genetic variants do not fully explain the heritability of complex traits, known as the missing heritability problem. To address this challenge and accurately control false positives while maximizing true associations, we propose two approaches involving linkage disequilibrium (LD) scores as covariates. We apply principal component analysis (PCA), one of the dimensionality reduction techniques, to control the False Discovery Rate (FDR) in the presence of high-dimensional covariates. This method not only provides a convenient interpretation of how multiple covariates in high dimensions affect the control of FDR but also offers higher statistical power compared to cases where covariates are not used. Furthermore, we aim to investigate how covariates contribute to increasing the statistical power through various simulation experiments, comparing the results with real data examples to derive better interpretations. Using real-world datasets, including GWAS with Body Mass Index (BMI) as the phenotype, we evaluate the performance of our proposed approaches. By incorporating LD scores as covariates in FDR-controlled GWAS analyzes, we demonstrate their effectiveness in selecting informative LD scores and improving the identification of significant SNPs. Our methods alleviate computational burden and enhance interpretability while retaining essential information from LD scores. In general, our study contributes to the advancement of statistical methods in GWAS and provides practical guidance for researchers looking to improve the precision of genetic association analyses.

stat.ME↗

arXiv2025

lsirm12pl: An R package for latent space item response modeling

The item response model in latent space (LSIRM; Jeon et al., 2021) uncovers unobserved interactions between respondents and items in the item response data by embedding both in a shared latent metric space. The R package lsirm12pl implements Bayesian estimation of the LSIRM and its extensions for various response types, base model specifications, and missing data handling. Furthermore, lsirm12pl package provides methods to improve model utilization and interpretation, such as clustering item positions on an estimated interaction map. The package also offers convenient summary and plotting options to evaluate and process the estimated results. In this paper, we provide an overview of the LSIRM's methodological foundation and describe several extensions included in the package. We then demonstrate the use of the package with real data examples contained within it.

stat.ME↗

arXiv2025

VR-Pipe: Streamlining Hardware Graphics Pipeline for Volume Rendering

Graphics rendering that builds on machine learning and radiance fields is gaining significant attention due to its outstanding quality and speed in generating photorealistic images from novel viewpoints. However, prior work has primarily focused on evaluating its performance through software-based rendering on programmable shader cores, leaving its performance when exploiting fixed-function graphics units largely unexplored. In this paper, we investigate the performance implications of performing radiance field rendering on the hardware graphics pipeline. In doing so, we implement the state-of-the-art radiance field method, 3D Gaussian splatting, using graphics APIs and evaluate it across synthetic and real-world scenes on today's graphics hardware. Based on our analysis, we present VR-Pipe, which seamlessly integrates two innovations into graphics hardware to streamline the hardware pipeline for volume rendering, such as radiance field methods. First, we introduce native hardware support for early termination by repurposing existing special-purpose hardware in modern GPUs. Second, we propose multi-granular tile binning with quad merging, which opportunistically blends fragments in shader cores before passing them to fixed-function blending units. Our evaluation shows that VR-Pipe greatly improves rendering performance, achieving up to a 2.78x speedup over the conventional graphics pipeline with negligible hardware overhead.

cs.GR↗

arXiv2024

General linear hypothesis testing of high-dimensional mean vectors with unequal covariance matrices based on random integration

This paper is devoted to the study of the general linear hypothesis testing (GLHT) problem of multi-sample high-dimensional mean vectors. For the GLHT problem, we introduce a test statistic based on $L^2$-norm and random integration method, and deduce the asymptotic distribution of the statistic under given conditions. Finally, the potential advantages of our test statistics are verified by numerical simulation studies and examples.

math.ST↗

arXiv2023

False Discovery Rate Controlling Procedures with BLOSUM62 substitution matrix and their application to HIV Data

Identifying significant sites in sequence data and analogous data is of fundamental importance in many biological fields. Fisher's exact test is a popular technique, however this approach to sparse count data is not appropriate due to conservative decisions. Since count data in HIV data are typically very sparse, it is crucial to use additional information to statistical models to improve testing power. In order to develop new approaches to incorporate biological information in the false discovery controlling procedure, we propose two models: one based on the empirical Bayes model under independence of amino acids and the other uses pairwise associations of amino acids based on Markov random field with on the BLOSUM62 substitution matrix. We apply the proposed methods to HIV data and identify significant sites incorporating BLOSUM62 matrix while the traditional method based on Fisher's test does not discover any site. These newly developed methods have the potential to handle many biological problems in the studies of vaccine and drug trials and phenotype studies.

stat.ME↗

arXiv2023

High dimensional discriminant rules with shrinkage estimators of the covariance matrix and mean vector

Linear discriminant analysis (LDA) is a typical method for classification problems with large dimensions and small samples. There are various types of LDA methods that are based on the different types of estimators for the covariance matrices and mean vectors. In this paper, we consider shrinkage methods based on a non-parametric approach. For the precision matrix, methods based on the sparsity structure or data splitting are examined. Regarding the estimation of mean vectors, Non-parametric Empirical Bayes (NPEB) methods and Non-parametric Maximum Likelihood Estimation (NPMLE) methods, also known as f-modeling and g-modeling, respectively, are adopted. The performance of linear discriminant rules based on combined estimation strategies of the covariance matrix and mean vectors are analyzed in this study. Particularly, the study presents a theoretical result on the performance of the NPEB method and compares it with previous studies. Simulation studies with various covariance matrices and mean vector structures are conducted to evaluate the methods discussed in this paper. Furthermore, real data examples such as gene expressions and EEG data are also presented

stat.ME↗

arXiv2021

Efficient Integration of Aggregate Data and Individual Patient Data in One-Way Mixed Models

Often both Aggregate Data (AD) studies and Individual Patient Data (IPD) studies are available for specific treatments. Combining these two sources of data could improve the overall meta-analytic estimates of treatment effects. Moreover, often for some studies with AD, the associated IPD maybe available, albeit at some extra effort or cost to the analyst. We propose a method for combining treatment effects across trials when the response is from the exponential family of distribution and hence a generalized linear model structure can be used. We consider the case when treatment effects are fixed and common across studies. Using the proposed combination method, we evaluate the wisdom of choosing AD when IPD is available by studying the relative efficiency of analyzing all IPD studies versus combining various percentages of AD and IPD studies. For many different models design constraints under which the AD estimators are the IPD estimators, and hence fully efficient, are known. For such models we advocate a selection procedure that chooses AD studies over IPD studies in a manner that force least departure from design constraints and hence ensures a fully efficient combined AD and IPD estimator.

stat.ME↗

arXiv2020

Dynamic Humanoid Locomotion over Uneven Terrain With Streamlined Perception-Control Pipeline

Although bipedal locomotion provides the ability to traverse unstructured environments, it requires careful planning and control to safely walk across without falling. This poses an integrated challenge for the robot to perceive, plan, and control its movements, especially with dynamic motions where the robot may have to adapt its swing-leg trajectory onthe-fly in order to safely place its foot on the uneven terrain. In this paper we present an efficient geometric footstep planner and the corresponding walking controller that enables a humanoid robot to dynamically walk across uneven terrain at speeds up to 0.3 m/s. As dynamic locomotion, we refer first to the continuous walking motion without stopping, and second to the on-the-fly replanning of the landing footstep position in middle of the swing phase during the robot gait cycle. This is mainly achieved through the streamlined integration between an efficient sampling-based planner and robust walking controller. The footstep planner is able to generate feasible footsteps within 5 milliseconds, and the controller is able to generate a new corresponding swing leg trajectory as well as the wholebody motion to dynamically balance the robot to the newly updated footsteps. The proposed perception-control pipeline is evaluated and demonstrated with real experiments using a fullscale humanoid to traverse across uneven terrains featured by static stepping stones, dynamically movable stepping stone, or narrow path.

cs.RO↗

arXiv2019

Note on Mean Vector Testing for High-Dimensional Dependent Observations

For the mean vector test in high dimension, Ayyala et al.(2017,153:136-155) proposed new test statistics when the observational vectors are M dependent. Under certain conditions, the test statistics for one-same and two-sample cases were shown to be asymptotically normal. While the test statistics and the asymptotic results are valid, some parts of the proof of asymptotic normality need to be corrected. In this work, we provide corrections to the proofs of their main theorems. We also note a few minor discrepancies in calculations in the publication.

math.ST↗