Search arXiv⌕ Search

arXiv subjects

Jinfeng Xu

Publications and source records attributed to Jinfeng Xu.

86 records · Page 5Linked to original sources
arXiv2024

MENTOR: Multi-level Self-supervised Learning for Multimodal Recommendation

With the increasing multimedia information, multimodal recommendation has received extensive attention. It utilizes multimodal information to alleviate the data sparsity problem in recommendation systems, thus improving recommendation accuracy. However, the reliance on labeled data severely limits the performance of multimodal recommendation models. Recently, self-supervised learning has been used in multimodal recommendations to mitigate the label sparsity problem. Nevertheless, the state-of-the-art methods cannot avoid the modality noise when aligning multimodal information due to the large differences in the distributions of different modalities. To this end, we propose a Multi-level sElf-supervised learNing for mulTimOdal Recommendation (MENTOR) method to address the label sparsity problem and the modality alignment problem. Specifically, MENTOR first enhances the specific features of each modality using the graph convolutional network (GCN) and fuses the visual and textual modalities. It then enhances the item representation via the item semantic graph for all modalities, including the fused modality. Then, it introduces two multilevel self-supervised tasks: the multilevel cross-modal alignment task and the general feature enhancement task. The multilevel cross-modal alignment task aligns each modality under the guidance of the ID embedding from multiple levels while maintaining the historical interaction information. The general feature enhancement task enhances the general feature from both the graph and feature perspectives to improve the robustness of our model. Extensive experiments on three publicly available datasets demonstrate the effectiveness of our method. Our code is publicly available at https://github.com/Jinfeng-Xu/MENTOR.

cs.IR↗
arXiv2023

Dual-Reference Source-Free Active Domain Adaptation for Nasopharyngeal Carcinoma Tumor Segmentation across Multiple Hospitals

Nasopharyngeal carcinoma (NPC) is a prevalent and clinically significant malignancy that predominantly impacts the head and neck area. Precise delineation of the Gross Tumor Volume (GTV) plays a pivotal role in ensuring effective radiotherapy for NPC. Despite recent methods that have achieved promising results on GTV segmentation, they are still limited by lacking carefully-annotated data and hard-to-access data from multiple hospitals in clinical practice. Although some unsupervised domain adaptation (UDA) has been proposed to alleviate this problem, unconditionally mapping the distribution distorts the underlying structural information, leading to inferior performance. To address this challenge, we devise a novel Sourece-Free Active Domain Adaptation (SFADA) framework to facilitate domain adaptation for the GTV segmentation task. Specifically, we design a dual reference strategy to select domain-invariant and domain-specific representative samples from a specific target domain for annotation and model fine-tuning without relying on source-domain data. Our approach not only ensures data privacy but also reduces the workload for oncologists as it just requires annotating a few representative samples from the target domain and does not need to access the source data. We collect a large-scale clinical dataset comprising 1057 NPC patients from five hospitals to validate our approach. Experimental results show that our method outperforms the UDA methods and achieves comparable results to the fully supervised upper bound, even with few annotations, highlighting the significant medical utility of our approach. In addition, there is no public dataset about multi-center NPC segmentation, we will release code and dataset for future research.

cs.CV↗
arXiv2023

Causal inference with outcome dependent sampling and mismeasured outcome

Outcome-dependent sampling designs are extensively utilized in various scientific disciplines, including epidemiology, ecology, and economics, with retrospective case-control studies being specific examples of such designs. Additionally, if the outcome used for sample selection is also mismeasured, then it is even more challenging to estimate the average treatment effect (ATE) accurately. To our knowledge, no existing method can address these two issues simultaneously. In this paper, we establish the identifiability of ATE and propose a novel method for estimating ATE in the context of generalized linear model. The estimator is shown to be consistent under some regularity conditions. To relax the model assumption, we also consider generalized additive model. We propose to estimate ATE using penalized B-splines and establish asymptotic properties for the proposed estimator. Our methods are evaluated through extensive simulation studies and the application to a dataset from the UK Biobank, with alcohol intake as the treatment and gout as the outcome.

stat.ME↗
arXiv2023

Strong Asymptotic Properties of Kernel Smoothing Estimation for NA Random Variables with Right Censoring

Most studies for negatively associated (NA) random variables consider the complete-data situation, which is actually a relatively ideal condition in practice. The paper relaxes this condition to the incomplete-data setting and considers kernel smoothing density and hazard function estimation in the presence of right censoring based on the Kaplan-Meier estimator. We establish the strong asymptotic properties for these two estimators to assess their asymptotic behavior and justify their practical use.

math.ST↗
arXiv2023

A degree-corrected Cox model for dynamic networks

Continuous time network data have been successfully modeled by multivariate counting processes, in which the intensity function is characterized by covariate information. However, degree heterogeneity has not been incorporated into the model which may lead to large biases for the estimation of homophily effects. In this paper, we propose a degree-corrected Cox network model to simultaneously analyze the dynamic degree heterogeneity and homophily effects for continuous time directed network data. Since each node has individual-specific in- and out-degree effects in the model, the dimension of the time-varying parameter vector grows with the number of nodes, which makes the estimation problem non-standard. We develop a local estimating equations approach to estimate unknown time-varying parameters, and establish consistency and asymptotic normality of the proposed estimators by using the powerful martingale process theories. We further propose test statistics to test for trend and degree heterogeneity in dynamic networks. Simulation studies are provided to assess the finite sample performance of the proposed method and a real data analysis is used to illustrate its practical utility.

math.ST↗
arXiv2022

CasFusionNet: A Cascaded Network for Point Cloud Semantic Scene Completion by Dense Feature Fusion

Semantic scene completion (SSC) aims to complete a partial 3D scene and predict its semantics simultaneously. Most existing works adopt the voxel representations, thus suffering from the growth of memory and computation cost as the voxel resolution increases. Though a few works attempt to solve SSC from the perspective of 3D point clouds, they have not fully exploited the correlation and complementarity between the two tasks of scene completion and semantic segmentation. In our work, we present CasFusionNet, a novel cascaded network for point cloud semantic scene completion by dense feature fusion. Specifically, we design (i) a global completion module (GCM) to produce an upsampled and completed but coarse point set, (ii) a semantic segmentation module (SSM) to predict the per-point semantic labels of the completed points generated by GCM, and (iii) a local refinement module (LRM) to further refine the coarse completed points and the associated labels from a local perspective. We organize the above three modules via dense feature fusion in each level, and cascade a total of four levels, where we also employ feature fusion between each level for sufficient information usage. Both quantitative and qualitative results on our compiled two point-based datasets validate the effectiveness and superiority of our CasFusionNet compared to state-of-the-art methods in terms of both scene completion and semantic segmentation. The codes and datasets are available at: https://github.com/JinfengX/CasFusionNet.

cs.CV↗
arXiv2022

Wilks' theorems in the $β$-model

Likelihood ratio tests and the Wilks theorems have been pivotal in statistics but have rarely been explored in network models with an increasing dimension. We are concerned here with likelihood ratio tests in the $β$-model for undirected graphs. For two growing dimensional null hypotheses including a specified null $H_0: β_i=β_i^0$ for $i=1,\ldots, r$ and a homogenous null $H_0: β_1=\cdots=β_r$, we reveal high dimensional Wilks' phenomena that the normalized log-likelihood ratio statistic, $[2\{\ell(\widehat{\boldsymbolβ}) - \ell(\widehat{\boldsymbolβ}^0)\} - r]/(2r)^{1/2}$, converges in distribution to the standard normal distribution as $r$ goes to infinity. Here, $\ell( \boldsymbolβ)$ is the log-likelihood function on the vector parameter $\boldsymbolβ=(β_1, \ldots, β_n)^\top$, $\widehat{\boldsymbolβ}$ is its maximum likelihood estimator (MLE) under the full parameter space, and $\widehat{\boldsymbolβ}^0$ is the restricted MLE under the null parameter space. For the corresponding fixed dimensional null $H_0: β_i=β_i^0$ for $i=1,\ldots, r$ and the homogenous null $H_0: β_1=\cdots=β_r$ with a fixed $r$, we establish Wilks type of results that $2\{\ell(\widehat{\boldsymbolβ}) - \ell(\widehat{\boldsymbolβ}^0)\}$ converges in distribution to a Chi-square distribution with respective $r$ and $r-1$ degrees of freedom, as the total number of parameters, $n$, goes to infinity. The Wilks type of results are further extended into a closely related Bradley--Terry model for paired comparisons, where we discover a different phenomenon that the log-likelihood ratio statistic under the fixed dimensional specified null asymptotically follows neither a Chi-square nor a rescaled Chi-square distribution. Simulation studies and an application to NBA data illustrate the theoretical results.

math.ST↗
arXiv2022

Improving Feature Extraction from Histopathological Images Through A Fine-tuning ImageNet Model

Due to lack of annotated pathological images, transfer learning has been the predominant approach in the field of digital pathology.Pre-trained neural networks based on ImageNet database are often used to extract "off the shelf" features, achieving great success in predicting tissue types, molecular features, and clinical outcomes, etc. We hypothesize that fine-tuning the pre-trained models using histopathological images could further improve feature extraction, and downstream prediction performance.We used 100,000 annotated HE image patches for colorectal cancer (CRC) to finetune a pretrained Xception model via a twostep approach.The features extracted from finetuned Xception (FTX2048) model and Imagepretrained (IMGNET2048) model were compared through: (1) tissue classification for HE images from CRC, same image type that was used for finetuning; (2) prediction of immunerelated gene expression and (3) gene mutations for lung adenocarcinoma (LUAD).Fivefold cross validation was used for model performance evaluation. The extracted features from the finetuned FTX2048 exhibited significantly higher accuracy for predicting tisue types of CRC compared to the off the shelf feature directly from Xception based on ImageNet database. Particularly, FTX2048 markedly improved the accuracy for stroma from 87% to 94%. Similarly, features from FTX2048 boosted the prediction of transcriptomic expression of immunerelated genesin LUAD. For the genes that had signigicant relationships with image fetures, the features fgrom the finetuned model imprroved the prediction for the majority of the genes. Inaddition, fetures from FTX2048 improved prediction of mutation for 5 out of 9 most frequently mutated genes in LUAD.

eess.IV↗
arXiv2021

Wilks' theorems in some exponential random graph models

We are concerned here with the likelihood ratio statistics in two exponential random graph models -- the $β$-model and the Bradley-Terry model, in which the degree sequence on an undirected graph and the out-degree sequence on a weighted directed graph are the exclusively sufficient statistics in the exponential-family distributions on graphs, respectively. We prove the Wilks type of theorems for some fixed and growing dimensional hypothesis testing problems. More specifically, under two fixed dimensional null hypotheses $H_0: β_i=β_i^0$ for $i=1,\ldots, r$ and $H_0: β_1=\ldots=β_r$, we show that $2[\ell(\widehat{\boldsymbolβ}) - \ell(\widehat{\boldsymbolβ}^0)]$ converges in distribution to a Chi-square distribution with the respective degrees of freedoms, $r$ and $r-1$, as the dimension $n$ of the full parameter space goes to infinity. Here, $\ell(\boldsymbolβ)$ is the log-likelihood function on the parameter $\boldsymbolβ$, $\widehat{\boldsymbolβ}$ is the MLE under the full parameter space, and $\widehat{\boldsymbolβ}^0$ is the restricted MLE under the null parameter space. For two increasing dimensional null hypotheses $H_0: β_i = β_i^0$ for $i=1, \ldots, n$ and $H_0: β_1=\ldots=β_r$ with $r/n \ge c$, we show that the normalized log-likelihood ratio statistics, $(2[\ell(\widehat{\boldsymbolβ}) - \ell(\boldsymbolβ^0)] -n)/(2n)^{1/2}$ and $(2[\ell(\widehat{\boldsymbolβ}) - \ell(\widehat{\boldsymbolβ}^0)] -r)/(2r)^{1/2}$, both converge in distribution to the standard normal distribution. Simulation studies and an application to NBA data illustrate the theoretical results.

math.ST↗
arXiv2021

Scalable Estimation and Inference with Large-scale or Online Survival Data

With the rapid development of data collection and aggregation technologies in many scientific disciplines, it is becoming increasingly ubiquitous to conduct large-scale or online regression to analyze real-world data and unveil real-world evidence. In such applications, it is often numerically challenging or sometimes infeasible to store the entire dataset in memory. Consequently, classical batch-based estimation methods that involve the entire dataset are less attractive or no longer applicable. Instead, recursive estimation methods such as stochastic gradient descent that process data points sequentially are more appealing, exhibiting both numerical convenience and memory efficiency. In this paper, for scalable estimation of large or online survival data, we propose a stochastic gradient descent method which recursively updates the estimates in an online manner as data points arrive sequentially in streams. Theoretical results such as asymptotic normality and estimation efficiency are established to justify its validity. Furthermore, to quantify the uncertainty associated with the proposed stochastic gradient descent estimator and facilitate statistical inference, we develop a scalable resampling strategy that specifically caters to the large-scale or online setting. Simulation studies and a real data application are also provided to assess its performance and illustrate its practical utility.

stat.ME↗
arXiv2020

Learning Latent Features with Pairwise Penalties in Low-Rank Matrix Completion

Low-rank matrix completion has achieved great success in many real-world data applications. A matrix factorization model that learns latent features is usually employed and, to improve prediction performance, the similarities between latent variables can be exploited by pairwise learning using the graph regularized matrix factorization (GRMF) method. However, existing GRMF approaches often use the squared loss to measure the pairwise differences, which may be overly influenced by dissimilar pairs and lead to inferior prediction. To fully empower pairwise learning for matrix completion, we propose a general optimization framework that allows a rich class of (non-)convex pairwise penalty functions. A new and efficient algorithm is developed to solve the proposed optimization problem, with a theoretical convergence guarantee under mild assumptions. In an important situation where the latent variables form a small number of subgroups, its statistical guarantee is also fully considered. In particular, we theoretically characterize the performance of the complexity-regularized maximum likelihood estimator, as a special case of our framework, which is shown to have smaller errors when compared to the standard matrix completion framework without pairwise penalties. We conduct extensive experiments on both synthetic and real datasets to demonstrate the superior performance of this general framework.

stat.ML↗
arXiv2017

On Scalable Inference with Stochastic Gradient Descent

In many applications involving large dataset or online updating, stochastic gradient descent (SGD) provides a scalable way to compute parameter estimates and has gained increasing popularity due to its numerical convenience and memory efficiency. While the asymptotic properties of SGD-based estimators have been established decades ago, statistical inference such as interval estimation remains much unexplored. The traditional resampling method such as the bootstrap is not computationally feasible since it requires to repeatedly draw independent samples from the entire dataset. The plug-in method is not applicable when there are no explicit formulas for the covariance matrix of the estimator. In this paper, we propose a scalable inferential procedure for stochastic gradient descent, which, upon the arrival of each observation, updates the SGD estimate as well as a large number of randomly perturbed SGD estimates. The proposed method is easy to implement in practice. We establish its theoretical properties for a general class of models that includes generalized linear models and quantile regression models as special cases. The finite-sample performance and numerical utility is evaluated by simulation studies and two real data applications.

stat.ML↗
arXiv2012

Grouped sparse paired comparisons in the Bradley-Terry model

In a wide class of paired comparisons, especially in the sports games, in which all subjects are divided into several groups, the intragroup comparisons are dense and the intergroup comparisons are sparse. Typical examples include the NFL regular season. Motivated by these situations, we propose group sparsity for paired comparisons and show the consistency and asymptotical normality of the maximum likelihood estimate in the Bradley-Terry model when the number of parameters goes to infinity in this paper. Simulations are carried out to illustrate the group sparsity and asymptotical results.

stat.OT↗