arXiv subjects

Younghoon Kim

Publications and source records attributed to Younghoon Kim.

At least 19 recordsLinked to original sources

arXiv2026

Analytic Standard Errors for Latent Gaussian Discrete-Valued Multivariate Time Series

Unlike their continuous-valued counterparts, there are no universally preferred methodologies for modeling discrete-valued time series. This is especially problematic in fields such as psychology and education, where repeated-measures data often take the form of count, dichotomous, and ordered categorical variables. To address the need for flexible methodology for analyzing discrete-valued time series data, a copula-style multivariate model defined through deterministic functions of a latent stationary Gaussian vector series has been proposed. This model has several promising features, including the ability to accommodate a wide variety of marginal distributions within the same model while also allowing for the most flexible autocorrelation structure possible. We extend this framework by deriving analytic standard errors to facilitate inference on the latent Gaussian dynamics. In so doing, we establish the joint asymptotic normality of estimators of the parameters governing the latent Gaussian series and the marginal distributions. The performance of these analytic standard errors is examined in a simulation study and an empirical application.

stat.ME↗

arXiv2026

Inference for High-Dimensional Sparse Spectral Precision Matrices

Statistical inference for the spectral precision matrix at a given frequency allows us to assess frequency-specific conditional relationships among the components of a stationary multivariate time series. Compared with classical analogues, inference in the spectral domain is more challenging due to the absence of closed-form asymptotic variance expressions for complex-valued estimators, the limited asymptotic theory in high-dimensional settings, and the presence of truncation and smoothing biases in finite samples. We construct a debiased complex graphical lasso estimator at any fixed frequency by leveraging the full likelihood structure of neighboring discrete Fourier transforms. Using asymptotic distributions for bilinear forms of stationary multivariate time series, we establish the joint asymptotic normality of the real and imaginary parts of the debiased estimator. Our main theoretical contributions include deriving a closed-form asymptotic covariance matrix for the real and imaginary parts, establishing a central limit theorem for bilinear forms of the underlying time series, and controlling smoothing and truncation biases in covariance estimation to ensure valid inference. Simulation studies demonstrate reliable coverage and improved statistical power relative to the benchmark, while maintaining false discovery rates near the nominal level. An application to real fMRI data further reveals distinct patterns of functional connectivity across selected frequencies.

stat.ME↗

arXiv2026

Latent community paths in VAR-type models via dynamic directed spectral co-clustering

This paper proposes a dynamic network framework for uncovering latent community paths in high-dimensional VAR-type models. By embedding a degree-corrected stochastic co-blockmodel (ScBM) into the transition matrices of VAR-type systems, we separate sending and receiving roles at the node level and summarize complex directional dependence in an interpretable low-dimensional form. Our method integrates directed spectral co-clustering with eigenvector smoothing to track how directional groups split, merge, or persist over time. This framework accommodates both periodic VAR (PVAR) models for cyclical seasonal evolution and generalized VHAR models for structural transitions across ordered dependence horizons. We establish non-asymptotic misclassification bounds for both procedures and provide supporting evidence through Monte Carlo experiments. Applications to U.S.\ nonfarm payrolls distinguish a recurrent business-centered core from more mobile, seasonally sensitive sectors. In global stock volatilities, the results reveal a compact U.S.-centered long-horizon block, a Europe-heavy developed core, and a more dynamic short-horizon reallocation of peripheral and bridge markets.

stat.ME↗

arXiv2025

Flex-MIG: Enabling Distributed Execution on MIG

GPU clusters in multi-tenant settings often suffer from underutilization, making GPU-sharing technologies essential for efficient resource use. Among them, NVIDIA Multi-Instance GPU (MIG) has gained traction for providing hardware-level isolation that enables concurrent workloads without interference. However, MIG's hardware rigidity and the conventional one-to-one allocation model jointly lead to severe fragmentation and cluster-wide underutilization. We present Flex-MIG, a software-only framework that replaces one-to-one with a one-to-many allocation model and enables host-shared-memory collectives across MIG instances without hardware modification. Flex-MIG eliminates drain-required reconfiguration, reduces fragmentation, and improves makespan by up to 17% across diverse traces, showing that rethinking MIG's operational model as a software-coordinated layer substantially improves cluster efficiency.

cs.DC↗

arXiv2025

Identification of Emotionally Stressful Periods Through Tracking Changes in Statistical Features of mHealth Data

Identifying the onset of emotional stress in older patients with mood disorders and chronic pain is crucial in mental health studies. To this end, studying the associations between passively sensed variables that measure human behaviors and self-reported stress levels collected from mobile devices is emerging. Existing algorithms rely on conventional change point detection (CPD) methods due to the nonstationary nature of the data. They also require explicit modeling of the associations between variables and output only discrete time points, which can lead to misinterpretation of stress onset timings. This is problematic when distributional shifts are complex, dependencies between variables are difficult to capture, and changes occur asynchronously across series with weak signals. In this study, we propose an algorithm that detects hotspots, defined as collections of time intervals during which statistical features of passive sensing variables and stress indicators shift, highlighting periods that require investigation. We first extend the moving sum (MOSUM) scheme to detect simultaneous changes both within and across series, and then define hotspots in two ways: using distance-based test statistics and confidence intervals. The proposed method tracks local changes in combined distributional features, enabling it to capture all types of simultaneous and asynchronous change. It does not require a specific functional relationship between series, and the results are expressed as intervals rather than as individual time points. We conduct simulations under varying signal strengths with mixed and asynchronous distributional shifts, where the proposed method outperforms benchmarks. Results on hotspot identification indicate that the two definitions are complementary. We further apply our method to ALACRITY Phase I data, analyzing hotspots from patients' stress levels and activity measures.

stat.ME↗

arXiv2025

Exact Coordinate Descent for High-Dimensional Regularized Huber Regression

This study develops a coordinate descent algorithm for high-dimensional Huber regression with an elastic-net penalty. Unlike existing gradient descent algorithms and coordinate descent methods based on second-order optimization, the proposed algorithm performs each coordinate update exactly, without approximation or gradient and Hessian computations. Consequently, it remains stable and converges rapidly even in the presence of highly correlated covariates or heavy-tailed noise. Building on these exact updates, adaptive variable screening rules and optimality-condition validation procedures are introduced to determine which coordinates to update at each iteration, substantially accelerating convergence in high-dimensional settings. Theoretical guarantees are established for both the convergence of the proposed algorithm and the validity of validation procedures. Extensive simulation studies under heavy-tailed noise and highly correlated designs, together with a real-data application, demonstrate that the proposed method is both accurate and computationally efficient in these challenging settings.

stat.ME↗

arXiv2025

Joint modeling and inference of multiple-subject high-dimensional sparse vector autoregressive models

The multiple-subject vector autoregression (multi-VAR) model captures heterogeneous network Granger causality across subjects by decomposing individual sparse VAR transition matrices into commonly shared and subject-unique paths. The model has been applied to characterize hidden shared and unique paths among subjects and has demonstrated performance compared to methods commonly used in psychology and neuroscience. Despite this innovation, the model suffers from using a weighted median for identifying the common effects, leading to statistical inefficiency as the convergence rates of the common and unique paths are determined by the least sparse subject and the smallest sample size across all subjects. We propose a new identifiability condition for the multi-VAR model based on a communication-efficient data integration framework. We show that this approach achieves convergence rates tailored to each subject's sparsity level and sample size. Furthermore, we develop hypothesis tests to assess the nullity and homogeneity of individual paths, using Wald-type test statistics constructed from individual debiased estimators. A test for the significance of the common paths can also be derived through the framework. Simulation studies under various heterogeneity scenarios and a real data application demonstrate the performance of the proposed method compared to existing benchmark across standard evaluation metrics.

stat.ME↗

arXiv2025

NetCAS: Dynamic Cache and Backend Device Management in Networked Environments

Modern storage systems often combine fast cache with slower backend devices to accelerate I/O. As performance gaps narrow, concurrently accessing both devices, rather than relying solely on cache hits, can improve throughput. However, in data centers, remote backend storage accessed over networks suffers from unpredictable contention, complicating this split. We present NetCAS, a framework that dynamically splits I/O between cache and backend devices based on real-time network feedback and a precomputed Perf Profile. Unlike traditional hit-rate-based policies, NetCAS adapts split ratios to workload configuration and networking performance. NetCAS employs a low-overhead batched round-robin scheduler to enforce splits, avoiding per-request costs. It achieves up to 174% higher performance than traditional caching in remote storage environments and outperforms converging schemes like Orthus by up to 3.5X under fluctuating network conditions.

cs.OS↗

arXiv2025

Latent Gaussian dynamic factor modeling and forecasting for multivariate count time series

This work considers estimation and forecasting in a multivariate, possibly high-dimensional count time series model constructed from a transformation of a latent Gaussian dynamic factor series. The estimation of the latent model parameters is based on second-order properties of the count and underlying Gaussian time series, yielding estimators of the underlying covariance matrices for which standard principal component analysis applies. Theoretical consistency results are established for the proposed estimation, building on certain concentration results for the models of the type considered. They also involve the memory of the latent Gaussian process, quantified through a spectral gap, shown to be suitably bounded as the model dimension increases, which is of independent interest. In addition, novel cross-validation schemes are suggested for model selection. The forecasting is carried out through a particle-based sequential Monte Carlo, leveraging Kalman filtering techniques. A simulation study and an application are also considered.

stat.ME↗

arXiv2025

A co-segmentation algorithm to predict emotional stress from passively sensed mHealth data

We develop a data-driven co-segmentation algorithm of passively sensed and self-reported active variables collected through smartphones to identify emotionally stressful states in middle-aged and older patients with mood disorders undergoing therapy, some of whom also have chronic pain. Our method leverages the association between the different types of time series. These data are typically non-stationary, with meaningful associations often occurring only over short time windows. Traditional machine learning (ML) methods, when applied globally on the entire time series, often fail to capture these time-varying local patterns. Our approach first segments the passive sensing variables by detecting their change points, then examines segment-specific associations with the active variable to identify co-segmented periods that exhibit distinct relationships between stress and passively sensed measures. We then use these periods to predict future emotional stress states using standard ML methods. By shifting the unit of analysis from individual time points to data-driven segments of time and allowing for different associations in different segments, our algorithm helps detect patterns that only exist within short-time windows. We apply our method to detect periods of stress in patient data collected during ALACRITY Phase I study. Our findings indicate that the data-driven segmentation algorithm identifies stress periods more accurately than traditional ML methods that do not incorporate segmentation.

stat.AP↗

arXiv2025

Dynamic spectral co-clustering of directed networks to unveil latent community paths in VAR-type models

Identifying network Granger causality in large vector autoregressive (VAR) models enhances explanatory power by capturing complex dependencies among variables. This study proposes a methodology that explores latent community structures to uncover underlying network dynamics, rather than relying on sparse coefficient estimation for network construction. A dynamic network framework embeds directed connectivity in the transition matrices of VAR-type models, allowing the tracking of evolving community structures over time, called seasons. To account for network directionality, degree-corrected stochastic co-block models are fitted for each season, then a combination of spectral co-clustering and singular vector smoothing is utilized to refine transitions between latent communities. Periodic VAR (PVAR) and vector heterogeneous autoregressive (VHAR) models are adopted as alternatives to conventional VAR models for dynamic network construction. Theoretical results establish the validity of the proposed methodology, while empirical analyses demonstrate its effectiveness in capturing both the cyclic evolution and transient trajectories of latent communities. The proposed approach is applied to US nonfarm payroll employment data and realized stock market volatility data. Spectral co-clustering of multi-layered directed networks, constructed from high-dimensional PVAR and VHAR representations, reveals rich and dynamic latent community structures.

stat.ME↗

arXiv2024

Group integrative dynamic factor models with application to multiple subject brain connectivity

This work introduces a novel framework for dynamic factor model-based group-level analysis of multiple subjects time series data, called GRoup Integrative DYnamic factor (GRIDY) models. The framework identifies and characterizes inter-subject similarities and differences between two pre-determined groups by considering a combination of group spatial information and individual temporal dynamics. Furthermore, it enables the identification of intra-subject similarities and differences over time by employing different model configurations for each subject. Methodologically, the framework combines a novel principal angle-based rank selection algorithm and a non-iterative integrative analysis framework. Inspired by simultaneous component analysis, this approach also reconstructs identifiable latent factor series with flexible covariance structures. The performance of the GRIDY models is evaluated through simulations conducted under various scenarios. An application is also presented to compare resting-state functional MRI data collected from multiple subjects in autism spectrum disorder and control groups.

stat.ME↗

arXiv2024

Mitigating Quantization Errors Due to Activation Spikes in GLU-Based LLMs

Modern large language models (LLMs) have established state-of-the-art performance through architectural improvements, but still require significant computational cost for inference. In an effort to reduce the inference cost, post-training quantization (PTQ) has become a popular approach, quantizing weights and activations to lower precision, such as INT8. In this paper, we reveal the challenges of activation quantization in GLU variants, which are widely used in feed-forward network (FFN) of modern LLMs, such as LLaMA family. The problem is that severe local quantization errors, caused by excessive magnitudes of activation in GLU variants, significantly degrade the performance of the quantized LLM. We denote these activations as activation spikes. Our further observations provide a systematic pattern of activation spikes: 1) The activation spikes occur in the FFN of specific layers, particularly in the early and late layers, 2) The activation spikes are dedicated to a couple of tokens, rather than being shared across a sequence. Based on our observations, we propose two empirical methods, Quantization-free Module (QFeM) and Quantization-free Prefix (QFeP), to isolate the activation spikes during quantization. Our extensive experiments validate the effectiveness of the proposed methods for the activation quantization, especially with coarse-grained scheme, of latest LLMs with GLU variants, including LLaMA-2/3, Mistral, Mixtral, SOLAR, and Gemma. In particular, our methods enhance the current alleviation techniques (e.g., SmoothQuant) that fail to control the activation spikes. Code is available at https://github.com/onnoo/activation-spikes.

cs.CL↗

arXiv2023

Structured Estimation of Heterogeneous Time Series

How best to model structurally heterogeneous processes is a foundational question in the social, health and behavioral sciences. Recently, Fisher et al., (2022) introduced the multi-VAR approach for simultaneously estimating multiple-subject multivariate time series characterized by common and individualizing features using penalized estimation. This approach differs from many popular modeling approaches for multiple-subject time series in that qualitative and quantitative differences in a large number of individual dynamics are well-accommodated. The current work extends the multi-VAR framework to include new adaptive weighting schemes that greatly improve estimation performance. In a small set of simulation studies we compare adaptive multi-VAR with these new penalty weights to common alternative estimators in terms of path recovery and bias. Furthermore, we provide toy examples and code demonstrating the utility of multi-VAR under different heterogeneity regimes using the multivar package for R (Fisher, 2022).

stat.ME↗

arXiv2023

Investigation of factors regarding the effects of COVID-19 pandemic on college students' depression by quantum annealer

Diverse cases regarding the impact, with its related factors, of the COVID-19 pandemic on mental health have been reported in previous studies. College student groups have been frequently selected as the target population in previous studies because they are easily affected by pandemics. In this study, multivariable datasets were collected from 751 college students based on the complex relationships between various mental health factors. We utilized quantum annealing (QA)-based feature selection algorithms that were executed by commercial D-Wave quantum computers to determine the changes in the relative importance of the associated factors before and after the pandemic. Multivariable linear regression (MLR) and XGBoost models were also applied to validate the QA-based algorithms. Based on the experimental results, we confirm that QA-based algorithms have comparable capabilities in factor analysis research to the MLR models that have been widely used in previous studies. Furthermore, the performance of the QA-based algorithms was validated through the important factor results from the algorithms. Pandemic-related factors (e.g., confidence in the social system) and psychological factors (e.g., decision-making in uncertain situations) were more important in post-pandemic conditions. We believe that our study will serve as a reference for researchers studying similar topics.

quant-ph↗

arXiv2023

Testing the Channels of Convolutional Neural Networks

Neural networks have complex structures, and thus it is hard to understand their inner workings and ensure correctness. To understand and debug convolutional neural networks (CNNs) we propose techniques for testing the channels of CNNs. We design FtGAN, an extension to GAN, that can generate test data with varying the intensity (i.e., sum of the neurons) of a channel of a target CNN. We also proposed a channel selection algorithm to find representative channels for testing. To efficiently inspect the target CNN's inference computations, we define unexpectedness score, which estimates how similar the inference computation of the test data is to that of the training data. We evaluated FtGAN with five public datasets and showed that our techniques successfully identify defective channels in five different CNN models.

cs.LG↗

arXiv2022

Multiple Instance Neural Networks Based on Sparse Attention for Cancer Detection using T-cell Receptor Sequences

Early detection of cancers has been much explored due to its paramount importance in biomedical fields. Among different types of data used to answer this biological question, studies based on T cell receptors (TCRs) are under recent spotlight due to the growing appreciation of the roles of the host immunity system in tumor biology. However, the one-to-many correspondence between a patient and multiple TCR sequences hinders researchers from simply adopting classical statistical/machine learning methods. There were recent attempts to model this type of data in the context of multiple instance learning (MIL). Despite the novel application of MIL to cancer detection using TCR sequences and the demonstrated adequate performance in several tumor types, there is still room for improvement, especially for certain cancer types. Furthermore, explainable neural network models are not fully investigated for this application. In this article, we propose multiple instance neural networks based on sparse attention (MINN-SA) to enhance the performance in cancer detection and explainability. The sparse attention structure drops out uninformative instances in each bag, achieving both interpretability and better predictive performance in combination with the skip connection. Our experiments show that MINN-SA yields the highest area under the ROC curve (AUC) scores on average measured across 10 different types of cancers, compared to existing MIL approaches. Moreover, we observe from the estimated attentions that MINN-SA can identify the TCRs that are specific for tumor antigens in the same T cell repertoire.

stat.ML↗

arXiv2022

Penalized Estimation and Forecasting of Multiple Subject Intensive Longitudinal Data

Intensive Longitudinal Data (ILD) is increasingly available to social and behavioral scientists. With this increased availability come new opportunities for modeling and predicting complex biological, behavioral, and physiological phenomena. Despite these new opportunities psychological researchers have not taken full advantage of promising opportunities inherent to this data, the potential to forecast psychological processes at the individual level. To address this gap in the literature we present a novel modeling framework that addresses a number of topical challenges and open questions in the psychological literature on modeling dynamic processes. First, how can we model and forecast ILD when the length of individual time series and the number of variables collected are roughly equivalent, or when time series lengths are shorter than what is typically required for time series analyses? Second, how can we best take advantage of the cross-sectional (between-person) information inherent to most ILD scenarios while acknowledging individuals differ both quantitatively (e.g. in parameter magnitude) and qualitatively (e.g. in structural dynamics)? Despite the acknowledged between-person heterogeneity in many psychological processes is it possible to leverage group-level information to support improved forecasting at the individual level? In the remainder of the manuscript, we attempt to address these and other pressing questions relevant to the forecasting of multiple-subject ILD.

stat.ME↗