arXiv subjects

Jianhua Hu

Publications and source records attributed to Jianhua Hu.

14 recordsLinked to original sources

arXiv2025

FedAPTA: Federated Multi-task Learning for Heterogeneous Devices with Adaptive Layer-wise Pruning and Task-aware Aggregation

Federated Learning (FL) has shown considerable promise in Machine Learning (ML) across numerous devices for privacy protection, efficient data utilization, and dynamic collaboration. However, mobile devices typically have limited and heterogeneous computational capabilities, and different devices may even have different tasks. This client heterogeneity is a major bottleneck hindering the practical application of FL. Existing work mainly focuses on mitigating FL's computation and communication overhead of a single task while overlooking the computing resource heterogeneity issue of different devices in FL. To tackle this, we design FedAPTA, a federated multi-task learning framework. FedAPTA overcomes computing resource heterogeneity through the developed layer-wise model pruning technique, which reduces local model size while considering both data and device heterogeneity. To aggregate structurally heterogeneous local models of different tasks, we introduce a heterogeneous model recovery strategy and a task-aware model aggregation method that enables the aggregation through infilling local model architecture with the shared global model and clustering local models according to their specific tasks. We deploy FedAPTA on a realistic FL platform and benchmark it against nine SOTA FL methods. The experimental outcomes demonstrate that the proposed FedAPTA considerably outperforms the state-of-the-art FL methods by up to 4.23\%. Our code is available at https://github.com/Zhenzovo/FedAPTA.

cs.DC↗

arXiv2025

Mesh of Spatiotemporal Optical Vortices with Programmable Intensity Nulls

Light carrying transverse orbital angular momentum (T-OAM) in the form of spatiotemporal optical vortices (STOVs) is opening new degrees of freedom for structured light manipulation. Such spatiotemporal wavepackets hold significant potential for optical trapping, analog optical computing, studying photonic symmetry and topology, among others. Up to now, synthesizing of such vortices is limited in one dimension, either in temporal or spatial domain. In this work, we propose and experimentally demonstrate a two-dimensional flexible mesh of spatiotemporal optical vortices (M-STOV) with programmable intensity nulls, and analyze their diffraction patterns for detection. Furthermore, we extend the spectral range of M-STOV via second-harmonic generation while examining the transfer of OAM in this nonlinear process. This study establishes a foundational framework for designing higher dimensional spatiotemporal vortex fields and promises a high-capacity information carrier based on ST optical vortices.

physics.optics↗

arXiv2025

Cascaded frequency conversion of highly charged femtosecond spatiotemporal optical vortices

The degrees of freedom inherent in spatiotemporal optical vortices (STOV's) afford intriguing opportunties to manipulate complex light fields for broad applications such as optical communication, light-matter interactions, particle manipulation, quantum optics, and electron acceleration in the relativistic regime. Unlike previous studies examining the second harmonic generation (SHG) of STOV's having an input topological charge (TC) l^{\left(ω\right)}=1, here we experimentally demonstrate cascaded second and third harmonic generation of STOV's to achieve more diverse wavelength selectivity via sum frequency generation (SFG), achieving unprecedented TC values up to l^{\left(ω\right)}=40. The large TC values are attributed to second (third) harmonic generation of an incident beam satisfying l^{\left(2ω\right)}=2l^{\left(ω\right)} (l^{\left(3ω\right)}=3l^{\left(ω\right)}). What is more, the wavelength of the generated STOV field was found to be tunable by controlling the position of the phase singularity in the frequency domain, and optimizing the nonlinear phase matching condition. Our experimental measurements extend the principle of conservation of the spatiotemporal topological charge to general nonlinear optical parametric processes, suggesting a fundamental approach to the production of STOV fields of arbitrarily large TC and at arbitrary visible wavelengths, and beyond (e.g., the ultraviolet range).

physics.optics↗

arXiv2025

Inverse Probability Weighting-based Mediation Analysis for Microbiome Data

Mediation analysis is an important tool for studying causal associations in biomedical and other scientific areas and has recently gained attention in microbiome studies. Using a microbiome study of acute myeloid leukemia (AML) patients, we investigate whether the effect of induction chemotherapy intensity levels on infection status is mediated by microbial taxa abundance. The unique characteristics of the microbial mediators -- high-dimensionality, zero-inflation, and dependence -- call for new methodological developments in mediation analysis. The presence of an exposure-induced mediator-outcome confounder, antibiotic use, further requires a delicate treatment in the analysis. To address these unique challenges in our motivating AML microbiome study, we propose a novel nonparametric identification formula for the interventional indirect effect (IIE), a recently developed measure for assessing mediation effects. We develop a corresponding estimation algorithm using the inverse probability weighting method. We also test the presence of mediation effects via constructing the standard normal bootstrap confidence intervals. Simulation studies demonstrate that the proposed method has good finite-sample performance in terms of IIE estimation accuracy and the type-I error rate and power of the corresponding tests. In the AML microbiome study, our findings suggest that the effect of induction chemotherapy intensity levels on infection is mainly mediated by patients' gut microbiome.

stat.AP↗

arXiv2025

The Tenth NTIRE 2025 Image Denoising Challenge Report

This paper presents an overview of the NTIRE 2025 Image Denoising Challenge (σ = 50), highlighting the proposed methodologies and corresponding results. The primary objective is to develop a network architecture capable of achieving high-quality denoising performance, quantitatively evaluated using PSNR, without constraints on computational complexity or model size. The task assumes independent additive white Gaussian noise (AWGN) with a fixed noise level of 50. A total of 290 participants registered for the challenge, with 20 teams successfully submitting valid results, providing insights into the current state-of-the-art in image denoising.

cs.CV↗

arXiv2025

TransST: Transfer Learning Embedded Spatial Factor Modeling of Spatial Transcriptomics Data

Background: Spatial transcriptomics have emerged as a powerful tool in biomedical research because of its ability to capture both the spatial contexts and abundance of the complete RNA transcript profile in organs of interest. However, limitations of the technology such as the relatively low resolution and comparatively insufficient sequencing depth make it difficult to reliably extract real biological signals from these data. To alleviate this challenge, we propose a novel transfer learning framework, referred to as TransST, to adaptively leverage the cell-labeled information from external sources in inferring cell-level heterogeneity of a target spatial transcriptomics data. Results: Applications in several real studies as well as a number of simulation settings show that our approach significantly improves existing techniques. For example, in the breast cancer study, TransST successfully identifies five biologically meaningful cell clusters, including the two subgroups of cancer in situ and invasive cancer; in addition, only TransST is able to separate the adipose tissues from the connective issues among all the studied methods. Conclusions: In summary, the proposed method TransST is both effective and robust in identifying cell subclusters and detecting corresponding driving biomarkers in spatial transcriptomics data.

q-bio.GN↗

arXiv2024

Zero-shot Generalizable Incremental Learning for Vision-Language Object Detection

This paper presents Incremental Vision-Language Object Detection (IVLOD), a novel learning task designed to incrementally adapt pre-trained Vision-Language Object Detection Models (VLODMs) to various specialized domains, while simultaneously preserving their zero-shot generalization capabilities for the generalized domain. To address this new challenge, we present the Zero-interference Reparameterizable Adaptation (ZiRa), a novel method that introduces Zero-interference Loss and reparameterization techniques to tackle IVLOD without incurring additional inference costs or a significant increase in memory usage. Comprehensive experiments on COCO and ODinW-13 datasets demonstrate that ZiRa effectively safeguards the zero-shot generalization ability of VLODMs while continuously adapting to new tasks. Specifically, after training on ODinW-13 datasets, ZiRa exhibits superior performance compared to CL-DETR and iDETR, boosting zero-shot generalizability by substantial 13.91 and 8.74 AP, respectively.Our code is available at https://github.com/JarintotionDin/ZiRaGroundingDINO.

cs.CV↗

arXiv2023

VisionFM: a Multi-Modal Multi-Task Vision Foundation Model for Generalist Ophthalmic Artificial Intelligence

We present VisionFM, a foundation model pre-trained with 3.4 million ophthalmic images from 560,457 individuals, covering a broad range of ophthalmic diseases, modalities, imaging devices, and demography. After pre-training, VisionFM provides a foundation to foster multiple ophthalmic artificial intelligence (AI) applications, such as disease screening and diagnosis, disease prognosis, subclassification of disease phenotype, and systemic biomarker and disease prediction, with each application enhanced with expert-level intelligence and accuracy. The generalist intelligence of VisionFM outperformed ophthalmologists with basic and intermediate levels in jointly diagnosing 12 common ophthalmic diseases. Evaluated on a new large-scale ophthalmic disease diagnosis benchmark database, as well as a new large-scale segmentation and detection benchmark database, VisionFM outperformed strong baseline deep neural networks. The ophthalmic image representations learned by VisionFM exhibited noteworthy explainability, and demonstrated strong generalizability to new ophthalmic modalities, disease spectrum, and imaging devices. As a foundation model, VisionFM has a large capacity to learn from diverse ophthalmic imaging data and disparate datasets. To be commensurate with this capacity, in addition to the real data used for pre-training, we also generated and leveraged synthetic ophthalmic imaging data. Experimental results revealed that synthetic data that passed visual Turing tests, can also enhance the representation learning capability of VisionFM, leading to substantial performance gains on downstream ophthalmic AI tasks. Beyond the ophthalmic AI applications developed, validated, and demonstrated in this work, substantial further applications can be achieved in an efficient and cost-effective manner using VisionFM as the foundation.

eess.IV↗

arXiv2023

Incremental Prototype Tuning for Class Incremental Learning

Class incremental learning(CIL) has attracted much attention, but most existing related works focus on fine-tuning the entire representation model, which inevitably results in much catastrophic forgetting. In the contrast, with a semantic-rich pre-trained representation model, parameter-additional-tuning (PAT) only changes very few parameters to learn new visual concepts. Recent studies have proved that PAT-based CIL can naturally avoid fighting against forgetting by replaying or distilling like most of the existing methods. However, we find that PAT-based CIL still faces serious semantic drift, the high-level forgetting problem caused by classifier learning bias at different learning phases, which significantly reduces the performance of PAT-based CIL. To address this problem, we propose Incremental Prototype Tuning (IPT), a simple but effective method that tunes category prototypes for classification and learning example prototypes to compensate for semantic drift. Extensive experiments demonstrate that our method can effectively compensate for semantic drift. Combined with well-pre-trained Vit backbones and other PAT methods, IPT surpasses the state-of-the-art baselines on mainstream incremental learning benchmarks.

cs.CV↗

arXiv2021

RZiMM-scRNA: A regularized zero-inflated mixture model framework for single-cell RNA-seq data

Applications of single-cell RNA sequencing in various biomedical research areas have been blooming. This new technology provides unprecedented opportunities to study disease heterogeneity at the cellular level. However, unique characteristics of scRNA-seq data, including large dimensionality, high dropout rates, and possibly batch effects, bring great difficulty into the analysis of such data. Not appropriately addressing these issues obstructs true scientific discovery. Herein, we propose a unified Regularized Zero-inflated Mixture Model framework designed for scRNA-seq data (RZiMM-scRNA) to simultaneously detect cell subgroups and identify gene differential expression based on a developed importance score, accounting for both dropouts and batch effects. We conduct extensive simulation studies in which we evaluate the performance of RZiMM-scRNA and compare it with several popular methods, including Seurat, SC3, K-Means, and Hierarchical Clustering. Simulation results show that RZiMM-scRNA demonstrates superior clustering performance and enhanced biomarker detection accuracy compared to alternative methods, especially when cell subgroups are less distinct, verifying the robustness of our method. Our empirical investigations focus on two brain tumor studies dealing with astrocytoma of various grades, including the most malignant of all brain tumors, glioblastoma multiforme (GBM). Our goal is to delineate cell heterogeneity and identify driving biomarkers associated with these tumors. Notably, RZiMM-scNRA successfully identifies a small group of oligodendrocyte cells which has drawn much attention in biomedical literature on brain cancers.

q-bio.GN↗

arXiv2021

Addressing patient heterogeneity in disease predictive model development

This paper addresses patient heterogeneity associated with prediction problems in biomedical applications. We propose a systematic hypothesis testing approach to determine the existence of patient subgroup structure and the number of subgroups in patient population if subgroups exist. A mixture of generalized linear models is considered to model the relationship between the disease outcome and patient characteristics and clinical factors, including targeted biomarker profiles. We construct a test statistic based on expectation maximization (EM) algorithm and derive its asymptotic distribution under the null hypothesis. An important computational advantage of the test is that the involved parameter estimates under the complex alternative hypothesis can be obtained through a small number of EM iterations, rather than optimizing the objective function. We demonstrate the finite sample performance of the proposed test in terms of type-I error rate and power, using extensive simulation studies. The applicability of the proposed method is illustrated through an application to a multi-center prostate cancer study.

stat.ME↗

arXiv2012

Estimation for an additive growth curve model with orthogonal design matrices

An additive growth curve model with orthogonal design matrices is proposed in which observations may have different profile forms. The proposed model allows us to fit data and then estimate parameters in a more parsimonious way than the traditional growth curve model. Two-stage generalized least-squares estimators for the regression coefficients are derived where a quadratic estimator for the covariance of observations is taken as the first-stage estimator. Consistency, asymptotic normality and asymptotic independence of these estimators are investigated. Simulation studies and a numerical example are given to illustrate the efficiency and parsimony of the proposed model for model specifications in the sense of minimizing Akaike's information criterion (AIC).

math.ST↗

arXiv2010

Sparse logistic principal components analysis for binary data

We develop a new principal components analysis (PCA) type dimension reduction method for binary data. Different from the standard PCA which is defined on the observed data, the proposed PCA is defined on the logit transform of the success probabilities of the binary observations. Sparsity is introduced to the principal component (PC) loading vectors for enhanced interpretability and more stable extraction of the principal components. Our sparse PCA is formulated as solving an optimization problem with a criterion function motivated from a penalized Bernoulli likelihood. A Majorization--Minimization algorithm is developed to efficiently solve the optimization problem. The effectiveness of the proposed sparse logistic PCA method is illustrated by application to a single nucleotide polymorphism data set and a simulation study.

stat.AP↗

arXiv2008

Asymptotic normality and consistency of a two-stage generalized least squares estimator in the growth curve model

Let $\mathbf{Y}=\mathbf{X}\boldsΘ\mathbf{Z}'+\bolds{\mathcal {E}}$ be the growth curve model with $\bolds{\mathcal{E}}$ distributed with mean $\mathbf{0}$ and covariance $\mathbf{I}_n\otimes\boldsΣ$, where $\boldsΘ$, $\boldsΣ$ are unknown matrices of parameters and $\mathbf{X}$, $\mathbf{Z}$ are known matrices. For the estimable parametric transformation of the form $\bolds γ=\mathbf{C}\boldsΘ\mathbf{D}'$ with given $\mathbf{C}$ and $\mathbf{D}$, the two-stage generalized least-squares estimator $\hat{\bolds γ}(\mathbf{Y})$ defined in (7) converges in probability to $\boldsγ$ as the sample size $n$ tends to infinity and, further, $\sqrt{n}[\hat{\boldsγ}(\mathbf{Y})-\bolds γ]$ converges in distribution to the multivariate normal distribution $\ma thcal{N}(\mathbf{0},(\mathbf{C}\mathbf{R}^{-1}\mathbf{C}')\otimes(\mat hbf{D}(\mathbf{Z}'\boldsΣ^{-1}\mathbf{Z})^{-1}\mathbf{D}'))$ under the condition that $\lim_{n\to\infty}\mathbf{X}'\mathbf{X}/n=\mathbf{R}$ for some positive definite matrix $\mathbf{R}$. Moreover, the unbiased and invariant quadratic estimator $\hat{\boldsΣ}(\mathbf{Y})$ defined in (6) is also proved to be consistent with the second-order parameter matrix $\boldsΣ$.

math.ST↗