Arrow Research search

Author name cluster

Yong Chen

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

36 papers
2 author rows

Possible papers

36

EAAI Journal 2026 Journal Article

A novel Brownian bridge diffusion-based generative inpainting algorithm for ancient murals

  • Yong Chen
  • Zhixin Fan
  • Shilong Zhang

Ancient murals, as one of the important types of ancient painting art, are an important carrier of human civilization and an important part of traditional culture. Existing deep learning algorithms for mural inpainting often lack constraints in global feature generation and structural information guidance, resulting in defects such as blurred edges and missing detailed textures in the inpainting murals. To address these limitations, this paper proposes a novel Brownian bridge diffusion-based generative inpainting algorithm for ancient murals. First, based on the physical constraints of the Brownian bridge process, we propose a diffusion process based on Brownian bridge to overcome the blindness of Gaussian noise addition in the traditional diffusion process. Subsequently, a line-drawing structure extraction module, which integrates HED-based edge embedding with a pyramid-structured autoencoder, is designed to provide detailed texture structures for models, solving the problem of missing detailed textures. Finally, a dual-prior guided inverse iterative inpainting module synergistically leverages both the Brownian bridge prior and the line-drawing guidance to enhance semantic coherence and generate detailed texture. Comparative experiments on a real Dunhuang mural dataset validate that the proposed method can effectively perform mural inpainting, and it delivers superior results in multi-angle evaluations compared to existing algorithms. Quantitative evaluations demonstrate that the proposed method outperforms state-of-the-art image inpainting algorithms across all metrics. Compared with the baseline diffusion model, our full approach improves Peak Signal-to-Noise Ratio (PSNR) by more than 35% and reduces the perceptual error (Learned Perceptual Image Patch Similarity, LPIPS) by over 53%, highlighting its superior performance. In addition, to verify the practical applicability of our algorithm, we designed an interactive ancient mural inpainting system with a visual interface that encapsulates the entire inpainting process. Although the proposed method achieves satisfactory mural inpainting performance, it still faces certain limitations. As the approach relies primarily on Red, Green, and Blue (RGB) visual information, the model struggles to capture the deeper cultural semantics — such as religious narratives and historical context — embedded in murals. Future work will explore multimodal learning and knowledge graph integration to enhance the model’s cultural awareness and semantic understanding in mural inpainting.

Details DOI

EAAI Journal 2026 Journal Article

An efficient foreign object detection model based on Transformer for high-speed railway catenary

  • Yong Chen
  • Jianyu Zhou
  • Chaoya Chen

Foreign object intrusion on the catenary can lead to power transmission failures or even major safety incidents. However, existing foreign object detection methods for catenaries often struggle with insufficient detection speed and suboptimal accuracy when faced with complex background interference and significant morphological variations of foreign objects. To overcome these issues, this paper proposes an efficient foreign object detection model based on Transformer for high-speed railway catenary. First, we design a lightweight backbone network called hybrid attention vision Transformer. This network combines the local perception of convolutional operations with the global modeling advantages of deformable large kernel attention, effectively capturing discriminative features of foreign objects amidst background interference. Second, we introduce a learnable gating mechanism and a feature enhancement module into the dynamic pixel decoder, enabling adaptive fusion of multi-scale features while maintaining accuracy and reducing computational overhead. Finally, we propose an improved Transformer decoder that integrates local self-attention with deformable kernels and a dynamic token merging strategy, enhancing both adaptability to catenary foreign objects and computational efficiency. Experimental results demonstrate the model’s effectiveness, achieving 85. 62% mean average precision (mAP) and 90. 53% mean intersection over union (mIoU), which significantly outperforms existing methods in accuracy. In terms of efficiency, our model contains only 34 million parameters and requires 51 giga floating-point operations (GFLOPs), yet achieves a high inference speed of 57. 5 frames per second (FPS). Our method improves foreign object detection in railway catenary systems by enhancing artificial intelligence and engineering collaboration, providing crucial support for high-speed rail applications.

Details DOI

I&C Journal 2026 Journal Article

Approximation algorithms for non-sequential star packing problems

  • Mengyuan Hu
  • An Zhang
  • Yong Chen
  • Mingyang Gong
  • Guohui Lin

For a positive integer k ≥ 1, a k-star ( k + -star, k − -star, respectively) is a connected graph containing a degree-ℓ vertex and ℓ degree-1 vertices, where ℓ = k ( ℓ ≥ k, 1 ≤ ℓ ≤ k, respectively). The k + -star packing problem is to cover as many vertices of an input graph G as possible using vertex-disjoint k + -stars in G; and given k > t ≥ 1, the k − / t -star packing problem is to cover as many vertices of G as possible using vertex-disjoint k − -stars but no t-stars in G. Both problems are NP-hard for any fixed k ≥ 2. We present a ( 1 + k 2 2 k + 1 ) - and a 3 2 -approximation algorithms for the k + -star packing problem when k ≥ 3 and k = 2, respectively, and a ( 1 + 1 t + 1 + 1 / k ) -approximation algorithm for the k − / t -star packing problem when k > t ≥ 2. They are all local search algorithms and they improve the best known approximation algorithms for the problems, respectively.

Details DOI

AAAI Conference 2026 Conference Paper

ARDiff: Anisotropic Residual Diffusion for Heterogeneous Graph Learning

  • Yong Chen
  • Li Li
  • Nannan Zong
  • Zhihui Liu
  • Song-Zhi Su

Learning representations on graphs is foundational for many downstream tasks, and its synergy with diffusion models has emerged as a promising direction. However, diffusion-based methods for heterogeneous graphs remain underexplored, confronting two principal challenges: (1) The presence of noise and structural heterogeneity in graphs makes it challenging to accurately capture semantic transitions among diverse relation types. (2) The isotropic Gaussian noise used in forward diffusion fails to reflect graphs' inherent semantics and structural anisotropy. To address these, we propose ARDiff, a novel framework that integrates residual diffusion with anisotropic noise for heterogeneous graph learning. Specifically, we propose a semantic residual diffusion mechanism that progressively refines node embeddings by orchestrating transitions from low-semantic (high-noise) to high-semantic (low-noise) relational contexts, thus enabling step-wise distillation of task-relevant information. In addition, to address the limitations of conventional diffusion, we introduce an anisotropic diffusion strategy: in the forward process, noise injection is oriented by structural and semantic priors; in the denoising step, a conditional diffusion mechanism is guided by a random walk encoding, enhancing both topological consistency and semantic alignment. Extensive evaluation on heterogeneous graph datasets demonstrates that ARDiff significantly surpasses current leading methods in link prediction and node classification, setting a new paradigm and benchmark in heterogeneous graph representation learning.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Codebook-Centric Deep Hashing: End-to-End Joint Learning of Semantic Hash Centers and Neural Hash Function

  • Shuo Yin
  • Zhiyuan Yin
  • Yuqing Hou
  • Rui Liu
  • Yong Chen
  • Dell Zhang

Hash center-based deep hashing methods improve upon pairwise or triplet-based approaches by assigning fixed hash centers to each class as learning targets, thereby avoiding the inefficiency of local similarity optimization. However, random center initialization often disregards inter-class semantic relationships. While existing two-stage methods mitigate this by first refining hash centers with semantics and then training the hash function, they introduce additional complexity, computational overhead, and suboptimal performance due to stage-wise discrepancies. To address these limitations, we propose Center-Reassigned Hashing (CRH), an end-to-end framework that dynamically reassigns hash centers from a preset codebook while jointly optimizing the hash function. Unlike previous methods, CRH adapts hash centers to the data distribution without explicit center optimization phases, enabling seamless integration of semantic relationships into the learning process. Furthermore, a multi-head mechanism enhances the representational capacity of hash centers, capturing richer semantic structures. Extensive experiments on three benchmarks demonstrate that CRH learns semantically meaningful hash centers and outperforms state-of-the-art deep hashing methods in retrieval tasks.

PDF Details DOI

AAAI Conference 2026 Conference Paper

FIXME: Towards End-to-End Benchmarking of LLM-Aided Design Verification

  • Gwok-Waa Wan
  • SamZaak Wong
  • Shengchu Su
  • Chenxu Niu
  • Ning Wang
  • Xinlai Wan
  • Qixiang Chen
  • Mengnv Xing

We introduce FIXME, the first end-to-end and large-scale benchmark for evaluating Large Language Models (LLMs) in hardware design functional verification (FV). Comprising 747 tasks derived from real-world hardware designs, FIXME spans five core FV sub-sets: specification comprehension, reference model generation, testbench generation, assertion design, and RTL debugging. To ensure high data quality, we developed an AI-human collaborative framework for agile data curation and annotation. This process resulted in 25,000 lines of verified RTL, 35,000 lines of enhanced testbenches, and over 1,200 SystemVerilog Assertions. Furthermore, through expert-guided optimization within the multi-agent aided flow, we achieved a remarkable 45.57% improvement in average functional coverage, underscoring the benchmark's robustness. Through evaluation of state-of-the-art LLMs like GPT-4.1, FIXME identifies key limitations and provides actionable insights, advancing the potential of LLM-driven automation in hardware design functional verification.

PDF Details DOI

EAAI Journal 2026 Journal Article

Time-varying formation robust control for connected nonlinear dynamic unmanned aerial vehicle systems

  • Shen Tang
  • Meng Li
  • Yong Chen
  • Yuyang Wu

This article investigates the problem of time-varying formation tracking control of connected nonlinear unmanned aerial vehicle systems (UAVSs) with dynamic external disturbances and dynamic directed topologies. A robust H∞ control strategy on the basis of time-varying delayed neighbor information is proposed. First, a dynamic model of connected Lipschitz nonlinear UAVSs with external disturbances and dynamic directed topologies is proposed. Second, a control algorithm that uses time-varying delayed neighbor information is developed to achieve formation tracking. Moreover, a variable transformation method is proposed, which transforms the time-varying formation-tracking problem into a lower-dimensional asymptotic-stability problem. Third, a new Lyapunov-Krasovskii function based on the free-weighting matrix method and derivative convex combinatorial lemma is proposed to analyze the stability of UAVSs. The sufficient condition for determining the control parameters is obtained by solving the linear matrix inequalities (LMIs). Lastly, the effectiveness of the proposed method is verified by numerical simulation.

Details DOI

AAAI Conference 2026 Conference Paper

TokenPowerBench: Benchmarking the Power Consumption of LLM Inference

  • Chenxu Niu
  • Wei Zhang
  • Jie Li
  • Yongjian Zhao
  • Tongyang Wang
  • Xi Wang
  • Yong Chen

Large language model (LLM) services now answer billions of queries per day, and industry reports show that inference, not training, accounts for more than 90% of total power consumption. However, existing benchmarks focus on either training/fine-tuning or performance of inference and provide little support for power consumption measurement and analysis of inference. We introduce TokenPowerBench, the first lightweight and extensible benchmark designed for LLM-inference power consumption studies. The benchmark combines a declarative configuration interface covering model choice, prompt set, and inference engine, a measurement layer that captures GPU-, node-, and system-level power without specialized power meters, and a phase-aligned metrics pipeline that attributes energy to the prefill and decode stages of every request. These elements make it straightforward to explore the power consumed by an LLM inference run; furthermore, by varying batch size, context length, parallelism strategy and quantization, users can quickly assess how each setting affects joules per token and other energy-efficiency metrics. We evaluate TokenPowerBench on four of the most widely used model series (Llama, Falcon, Qwen, and Mistral). Our experiments cover from 1 billion parameters up to the frontier-scale Llama3-405B model. Furthermore, we release TokenPowerBench as open source to help users to measure power consumption, forecast operating expenses, and meet sustainability targets when deploying LLM services.

PDF Details DOI

AAAI Conference 2026 Conference Paper

VBF++: Variational Bayesian Fusion with Context-Aware Priors and Recommendation-Guided Adversarial Refinement for Multimodal Video Recommendation

  • Ziyi Cao
  • Rui Liu
  • Yong Chen

Multimodal video recommendation systems face fundamental challenges in determining optimal fusion strategies across diverse content types and user preferences. Existing methods suffer from two critical limitations: (1) their fusion strategies are guided by context-agnostic priors that ignore the semantic structure of content, assuming the same simple distribution (typically a standard multivariate Gaussian prior) governs optimal fusion for all video types, and (2) their optimization objectives, particularly the Evidence Lower Bound (ELBO), are misaligned with the final recommendation goal, optimizing for feature reconstruction rather than ranking performance. To address these fundamental issues, this work proposes VBF++, a novel framework that introduces context-aware structured priors and recommendation-guided adversarial refinement. First, the method designs context-aware priors that learn cluster-specific distributions based on video semantic categories, replacing uninformative priors with structured, content-aware prior distributions. Second, it introduces a Recommendation-Guided Adversarial Refinement (RAR) paradigm that explicitly steers the learning process towards generating recommendation-optimal fusion strategies, resolving the objective misalignment inherent in variational learning. Enhanced with domain-adaptive meta-learning, extensive experiments on three real-world datasets demonstrate consistent improvements of 4.7-8.3 percent in Precision@10 over state-of-the-art methods. Analysis reveals that learned fusion strategies exhibit semantically meaningful patterns, prioritizing visual features for action content, acoustic information for music videos, and textual descriptions for documentary material.

PDF Details DOI

I&C Journal 2025 Journal Article

Approximation algorithms for the maximum path cover problem using long paths

  • Mingyang Gong
  • Yong Chen
  • Zhi-Zhong Chen
  • Guohui Lin
  • Bing Su
  • Lusheng Wang

The problem studied in this paper is to find a collection of vertex-disjoint paths in a given graph G = ( V, E ) such that each path has length at least k, called a long path, and the total number of edges on these paths is maximized. The problem is NP-hard for any fixed k or when k is part of the input, by a reduction from the Hamiltonian path problem. Berman and Karpinski presented a 7/6-approximation algorithm for k = 1, but for a general k ≥ 2, there is no approximation algorithm directly for the problem. We present the first local search ( 0. 4394 k + O ( 1 ) ) -approximation algorithm for any fixed k ≥ 1, and a 1. 4254-approximation algorithm for k = 2 built on top of a maximum triangle-free path-cycle cover.

Details DOI

EAAI Journal 2025 Journal Article

Cost-effective and real-time landslide monitoring method based on ultra-wideband using ultra-wideband transformer neural network

  • Yu Si
  • Zhaofeng He
  • Fan Zhang
  • Xiaoyun Sun
  • Yong Chen
  • Haiqing Zheng

Landslides rank among the most destructive natural phenomena, posing substantial risks to human safety, infrastructure, and ecological systems. Their frequent occurrence in topographically complex regions demands urgent development in real-time monitoring solutions. Current monitoring methodologies, however, are constrained by prohibitive costs, limited temporal resolution, and high-power consumption. These factors create substantial implementation barriers to implementing landslide monitoring systems. To address these limitations, this study proposes an economical real-time monitoring method leveraging ultra-wideband (UWB) technology for landslide detection. The implementation of a dual-Microcontroller Unit (MCU) distributed hardware architecture enables high-accuracy ranging capabilities and high real-time performance. To enhance the spatial resolution of UWB systems in landslide monitoring, we propose an optimized sensor deployment structure and a novel deep learning architecture called Ultra-wideband Transformer (UWBformer). This network utilizes differential UWB-ranging data to predict spatial displacement at monitoring locations, specifically the displacement distance, horizontal angle, and pitch angle. UWBformer incorporates a spatial multi-head attention mechanism and a dual-channel architecture processing both time-domain and frequency-domain features. It is specifically designed to mitigate ranging error propagation and enhance prediction stability by focusing on relative distance changes rather than absolute ranging accuracy. Empirical results demonstrate UWBformer's superior performance in predicting displacement distance, horizontal angle, and pitch angle, outperforming the conventional Caffery-Taylor (C-T) localization approach and established deep learning benchmarks. Field tests incorporated 3 σ criterion and Kalman filtering alongside to pre-process raw measurements, thereby enhancing data stability. Comprehensive validation across field tests demonstrates UWBformer's capability to maintain accurate spatial displacement estimation under harsh environments.

Details DOI

JMLR Journal 2025 Journal Article

DisC2o-HD: Distributed causal inference with covariates shift for analyzing real-world high-dimensional data

  • Jiayi Tong
  • Jie Hu
  • George Hripcsak
  • Yang Ning
  • Yong Chen

High-dimensional healthcare data, such as electronic health records (EHR) data and claims data, present two primary challenges due to the large number of variables and the need to consolidate data from multiple clinical sites. The third key challenge is the potential existence of heterogeneity in terms of covariate shift. In this paper, we propose a distributed learning algorithm accounting for covariate shift to estimate the average treatment effect (ATE) for high-dimensional data, named DisC2o-HD. Leveraging the surrogate likelihood method, our method calibrates the estimates of the propensity score and outcome models to approximately attain the desired covariate balancing property, while accounting for the covariate shift across multiple clinical sites. We show that our distributed covariate balancing propensity score estimator can approximate the pooled estimator, which is obtained by pooling the data from multiple sites together. The proposed estimator remains consistent if either the propensity score model or the outcome regression model is correctly specified. The semiparametric efficiency bound is achieved when both the propensity score and the outcome models are correctly specified. We conduct simulation studies to demonstrate the performance of the proposed algorithm; additionally, we conduct an empirical study to present the readiness of implementation and validity. [abs] [ pdf ][ bib ] &copy JMLR 2025. ( edit, beta )

PDF Details

NeurIPS Conference 2025 Conference Paper

FAME: Adaptive Functional Attention with Expert Routing for Function-on-Function Regression

  • Yifei Gao
  • Yong Chen
  • Chen Zhang

Functional data play a pivotal role across science and engineering, yet their infinite-dimensional nature makes representation learning challenging. Conventional statistical models depend on pre-chosen basis expansions or kernels, limiting the flexibility of data-driven discovery, while many deep-learning pipelines treat functions as fixed-grid vectors, ignoring inherent continuity. In this paper, we introduce Functional Attention with a Mixture-of-Experts (FAME), an end-to-end, fully data-driven framework for function-on-function regression. FAME forms continuous attention by coupling a bidirectional neural controlled differential equation with MoE-driven vector fields to capture intra-functional continuity, and further fuses change to inter-functional dependencies via multi-head cross attention. Extensive experiments on synthetic and real-world functional regression benchmarks show that FAME achieves state-of-the-art accuracy and strong robustness to arbitrarily sampled discrete observations of functions.

PDF Details

AAAI Conference 2025 Conference Paper

FreeGen: Bridging Visual-Linguistic Discrepancies Towards Diffusion-based Pixel-level Data Synthesis

  • Wenzhuang Wang
  • Mingcan Ma
  • Yong Chen
  • Changqun Xia
  • Zhenbao Liang
  • Jia Li

Text-to-image diffusion model has inspired research into text-to-data synthesis without human intervention, where spatial attentions correlated with semantic entities in text prompts are primarily interpreted as pseudo-masks. However, these vannila attentions often deliver visual-linguistic discrepancies, in which the associations between image features and entity-level tokens are unstable and divergent, yielding inferior masks for realistic applications, especially in more practical open-vocabulary settings. To tackle this issue, we propose a novel text-guided self-driven generative paradigm, termed FreeGen, which addresses the discrepancies by recalibrating intrinsic visual-linguistic correlations and serves as a real-data-free method to automatically synthesize open-vocabulary pixel-level data for arbitrary entities. Specifically, we first learn an Attention Self-Rectification mechanism to reproject the inherent attention matrices to achieve robust semantic alignment, thereby obtaining class-discriminative masks. A Temporal Fluctuation Factor is present to assess mask quality based on its variation over uniform sampling timesteps, enabling the selection of reliable masks. These masks are then employed as self-supervised signals to support the learning of an Entity-level Grounding Decoder in a self-training manner, thus producing open-vocabulary segmentation results. Extensive experiments show that the existing segmenters trained on FreeGen narrow the performance gap with real data counterparts and remarkably outperform the state-of-the-art methods.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

GraphAD: Interaction Scene Graph for End-to-end Autonomous Driving

  • Yunpeng Zhang
  • Deheng Qian
  • Ding Li
  • Yifeng Pan
  • Yong Chen
  • Zhenbao Liang
  • Zhiyao Zhang
  • Yingzong Liu

Modeling complicated interactions among the ego-vehicle, road agents, and map elements has been a crucial part for safety-critical autonomous driving. Previous work on end-to-end autonomous driving relies on the attention mechanism to handle heterogeneous interactions, which fails to capture geometric priors and is also computationally intensive. In this paper, we propose the Interaction Scene Graph (ISG) as a unified method to model the interactions among the ego-vehicle, road agents, and map elements. With the representation of the ISG, the driving agents aggregate essential information from the most influential elements, including the road agents with potential collisions and the map elements to follow. Since a mass of unnecessary interactions are omitted, the more efficient scene-graph-based framework is able to focus on indispensable connections and leads to better performance. We evaluate the proposed method for end-to-end autonomous driving on the nuScenes dataset. Compared with strong baselines, our method significantly outperforms in full-stack driving tasks.

PDF Details DOI

EAAI Journal 2025 Journal Article

Low-light image enhancement network based on central difference convolution

  • Yong Chen
  • Shangming Chen
  • Huanlin Liu
  • Hangying Xiong
  • Yourui Zhang

Since the convolutional neural networks and transformers used in existing low-light image enhancement methods were prone to ignore high-frequency information, resulting in blurred details of the enhanced image, this affected the performance of computer vision tasks at night. Therefore, we propose a novel low-light image enhancement network based on central difference convolution (CDCLNet). This method uses traditional image processing methods to help the network extract high-frequency information. Specifically, firstly, in order to fully expose the hidden high-frequency details, the proposed method uses the multi-exposure strategy based on bright and dark masks to expose the image to different levels. Secondly, the complementary information between multi-exposure images is fused through the first-stage network. Finally, the second-stage network suppresses the amplified noise and enhances the details. In addition, We design a central difference convolution module (CDCM) with channel attention to adaptively extract gradient-level detailed features according to the need of the two-stage network. In order to make the network notice illumination non-uniformity, we propose a multi-scale feature attention module (MFAM), which extracts multi-scale features in each channel and generates channel-specific attention maps. Experiments on four public datasets show that the proposed method can enhance the details more effectively than mainstream methods, and achieves the highest structural similarity index on two paired datasets, with an average value of 0. 899.

Details DOI

EAAI Journal 2025 Journal Article

Prediction of submarine soil dredging difficulty scale in cutter suction dredger construction with clustering-based deep learning

  • Yong Chen
  • Qiubing Ren
  • Mingchao Li
  • Huijing Tian
  • Liang Qin
  • Dianchun Wu

Expensive marine geological surveys and subjective human assessments limit the accuracy of submarine soil information, making it challenging to predict dredging difficulty for cutter suction dredger (CSD) excavation. To this end, we propose a data-driven framework that comprises data preprocessing, unsupervised clustering, and time series prediction using CSD construction big data. First, a high-dimensional feature selection method is employed to identify key features significantly affecting cutter cutting torque (CCT) and winch swing torque (WST) from 256-dimensional construction data. Then, the K-means algorithm defines a dredging difficulty scale (DDS) for dividing dredging difficulty by clustering CCT and WST. Finally, a deep learning model integrating the convolutional neural network (CNN), long short-term memory (LSTM), and the attention mechanism is formulated. The CNN-LSTM-Attention model aims to predict CCT and WST in the context of multivariate time series and then map such predictions to different DDSs. The applicability of the proposed framework is validated using 14, 400 construction data collected from Tian Jing Hao CSD. Results show that four DDSs defined can effectively represent various dredging difficulty levels. The CNN-LSTM-Attention model achieves a high real-time prediction accuracy of 95. 83% for DDSs and maintains robust performance across various prediction steps, which outperform baseline models. The proposed framework provides a novel approach for dividing and predicting dredging difficulty without relying on soil information, helping operators to optimize operational instructions in advance.

Details DOI

ICML Conference 2025 Conference Paper

SAFER: A Calibrated Risk-Aware Multimodal Recommendation Model for Dynamic Treatment Regimes

  • Yishan Shen
  • Yuyang Ye 0002
  • Hui Xiong 0001
  • Yong Chen

Dynamic treatment regimes (DTRs) are critical to precision medicine, optimizing long-term outcomes through personalized, real-time decision-making in evolving clinical contexts, but require careful supervision for unsafe treatment risks. Existing efforts rely primarily on clinician-prescribed gold standards despite the absence of a known optimal strategy, and predominantly using structured EHR data without extracting valuable insights from clinical notes, limiting their reliability for treatment recommendations. In this work, we introduce SAFER, a calibrated risk-aware tabular-language recommendation framework for DTR that integrates both structured EHR and clinical notes, enabling them to learn from each other, and addresses inherent label uncertainty by assuming ambiguous optimal treatment solution for deceased patients. Moreover, SAFER employs conformal prediction to provide statistical guarantees, ensuring safe treatment recommendations while filtering out uncertain predictions. Experiments on two publicly available sepsis datasets demonstrate that SAFER outperforms state-of-the-art baselines across multiple recommendation metrics and counterfactual mortality rate, while offering robust formal assurances. These findings underscore SAFER’s potential as a trustworthy and theoretically grounded solution for high-stakes DTR applications.

Details

TMLR Journal 2024 Journal Article

A Multilinear Least-Squares Formulation for Sparse Tensor Canonical Correlation Analysis

  • Jun Yu
  • Zhaoming Kong
  • Kun Chen
  • Xin Zhang
  • Yong Chen
  • Lifang He

Tensor data are becoming important recently in various applications, e.g., image and video recognition, which pose new challenges for data modeling and analysis approaches, such as high-order relations of large complexity, varying data scale and gross noise. In this paper, we consider the problem of sparse canonical correlation analysis for arbitrary tensor data. Although several methods have been proposed for this task, there are still limitations hindering its practical applications. To this end, we present a general Sparse Tensor Canonical Correlation Analysis (gSTCCA) method from a multilinear least-squares perspective. Specifically, we formulate the problem as a constrained multilinear least-squares problem with tensor-structured sparsity regularization based on CANDECOMP/PARAFAC (CP) decomposition. Then we present a divide-and-conquer deflation approach to tackle the problem by successive rank-one tensor estimation of the residual tensors, where the overall model is broken up into a set of unconstrained linear least-squares problems that can be efficiently solved. Through extensive experiments conducted on five different datasets for recognition tasks, we demonstrate that the proposed method achieves promising performance compared to the SOTA vector- and tensor-based canonical correlation analysis methods in terms of classification accuracy, model sparsity, and robustness to missing and noisy data. The code is publicly available at https://github.com/junfish/gSTCCA.

PDF Details

I&C Journal 2024 Journal Article

Approximating the directed path partition problem

  • Yong Chen
  • Zhi-Zhong Chen
  • Curtis Kennedy
  • Guohui Lin
  • Yao Xu
  • An Zhang

Given a digraph G = ( V, E ), the k-path partition problem aims to find a minimum collection of vertex-disjoint directed paths, of order at most k, to cover all the vertices. The problem has various applications. Its special case on undirected graphs is NP-hard when k ≥ 3, and has received much study recently from the approximation algorithm perspective. However, the general problem on digraphs is seemingly untouched in the literature. We fill the gap with the first k / 2 -approximation algorithm, based on a novel concept of enlarging walk to minimize the number of singletons. Secondly, for k = 3, we define a second novel kind of enlarging walks to greedily reduce the number of 2-paths in the 3-path partition and propose an improved 13/9-approximation algorithm. Lastly, for any k ≥ 7, we present an improved ( k + 2 ) / 3 -approximation algorithm built on the maximum path-cycle cover followed by a careful 2-cycle elimination process.

Details DOI

JBHI Journal 2024 Journal Article

SSCFormer: Revisiting ConvNet-Transformer Hybrid Framework From Scale-Wise and Spatial-Channel-Aware Perspectives for Volumetric Medical Image Segmentation

  • Qinlan Xie
  • Yong Chen
  • Shenglin Liu
  • Xuesong Lu

Accurate and robust medical image segmentation is crucial for assisting disease diagnosis, making treatment plan, and monitoring disease progression. Adaptive to different scale variations and regions of interest is essential for high accuracy in automatic segmentation methods. Existing methods based on the U-shaped architecture respectively tackling intra- and inter-scale problem with a hierarchical encoder, however, are restricted by the scope of multi-scale modeling. In addition, global attention and scaling attention in regions of interest have not been appropriately adopted, especially for the salient features. To address these two issues, we propose a ConvNet-Transformer hybrid framework named SSCFormer for accurate and versatile medical image segmentation. The intra-scale ResInception and inter-scale transformer bridge are designed to collaboratively capture the intra- and inter-scale features, facilitating the interaction of small-scale disparity information at a single stage with large-scale from multiple stages. Global attention and scaling attention are cleverly integrated from a spatial-channel-aware perspective. The proposed SSCFormer is tested on four different medical image segmentation tasks. Comprehensive experimental results show that SSCFormer outperforms the current state-of-the-art methods.

Details DOI

AAAI Conference 2023 Conference Paper

Beyond Graph Convolutional Network: An Interpretable Regularizer-Centered Optimization Framework

  • Shiping Wang
  • Zhihao Wu
  • Yuhong Chen
  • Yong Chen

Graph convolutional networks (GCNs) have been attracting widespread attentions due to their encouraging performance and powerful generalizations. However, few work provide a general view to interpret various GCNs and guide GCNs' designs. In this paper, by revisiting the original GCN, we induce an interpretable regularizer-centerd optimization framework, in which by building appropriate regularizers we can interpret most GCNs, such as APPNP, JKNet, DAGNN, and GNN-LF/HF. Further, under the proposed framework, we devise a dual-regularizer graph convolutional network (dubbed tsGCN) to capture topological and semantic structures from graph data. Since the derived learning rule for tsGCN contains an inverse of a large matrix and thus is time-consuming, we leverage the Woodbury matrix identity and low-rank approximation tricks to successfully decrease the high computational complexity of computing infinite-order graph convolutions. Extensive experiments on eight public datasets demonstrate that tsGCN achieves superior performance against quite a few state-of-the-art competitors w.r.t. classification tasks.

PDF Details DOI

TCS Journal 2023 Journal Article

Complexity and approximation algorithms for two parallel dedicated machine scheduling with conflict constraints

  • An Zhang
  • Liang Zhang
  • Yong Chen
  • Guangting Chen
  • Xing Wang

We investigate two parallel dedicated machine scheduling with conflict constraints. The problem of minimizing the makespan has been shown to be NP-hard in the strong sense under the assumption that the processing sequence of jobs on one machine is given and fixed a priori. The problem without any fixed sequence was previously recognized as weakly NP-hard. In this paper, we first present a 9 5 -approximation algorithm for the problem with a fixed sequence. Then we show that the tight approximation ratios of the algorithm are 7 4 and 5 3 for two subproblems which remain strongly NP-hard. We also send an improved algorithm with approximation ratio 3 − 2 ≈ 1. 586 for one subproblem. Finally, we prove that the problem without any fixed sequence is actually strongly NP-hard, and design a 5 3 -approximation algorithm.

Details DOI

JMLR Journal 2023 Journal Article

Scalable high-dimensional Bayesian varying coefficient models with unknown within-subject covariance

  • Ray Bai
  • Mary R. Boland
  • Yong Chen

Nonparametric varying coefficient (NVC) models are useful for modeling time-varying effects on responses that are measured repeatedly for the same subjects. When the number of covariates is moderate or large, it is desirable to perform variable selection from the varying coefficient functions. However, existing methods for variable selection in NVC models either fail to account for within-subject correlations or require the practitioner to specify a parametric form for the correlation structure. In this paper, we introduce the nonparametric varying coefficient spike-and-slab lasso (NVC-SSL) for Bayesian high dimensional NVC models. Through the introduction of functional random effects, our method allows for flexible modeling of within-subject correlations without needing to specify a parametric covariance function. We further propose several scalable optimization and Markov chain Monte Carlo (MCMC) algorithms. For variable selection, we propose an Expectation Conditional Maximization (ECM) algorithm to rapidly obtain maximum a posteriori (MAP) estimates. Our ECM algorithm scales linearly in the total number of observations $N$ and the number of covariates $p$. For uncertainty quantification, we introduce an approximate MCMC algorithm that also scales linearly in both $N$ and $p$. We demonstrate the scalability, variable selection performance, and inferential capabilities of our method through simulations and a real data application. These algorithms are implemented in the publicly available R package NVCSSL on the Comprehensive R Archive Network. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2023. ( edit, beta )

PDF Details

TIST Journal 2022 Journal Article

Federated Multi-view Learning for Private Medical Data Integration and Analysis

  • Sicong Che
  • Zhaoming Kong
  • Hao Peng
  • Lichao Sun
  • Alex Leow
  • Yong Chen
  • Lifang He

Along with the rapid expansion of information technology and digitalization of health data, there is an increasing concern on maintaining data privacy while garnering the benefits in the medical field. Two critical challenges are identified: First, medical data is naturally distributed across multiple local sites, making it difficult to collectively train machine learning models without data leakage. Second, in medical applications, data are often collected from different sources and views, resulting in heterogeneity and complexity that requires reconciliation. In this article, we present a generic Federated Multi-view Learning (FedMV) framework for multi-view data leakage prevention. Specifically, we apply this framework to two types of problems based on local data availability: Vertical Federated Multi-view Learning (V-FedMV) and Horizontal Federated Multi-view Learning (H-FedMV). We experimented with real-world keyboard data collected from BiAffect study. Our results demonstrated that the proposed approach can make full use of multi-view data in a privacy-preserving way, and both V-FedMV and H-FedMV perform better than their single-view and pairwise counterparts. Besides, the framework can be easily adapted to deal with multi-view sequential data. We have developed a sequential model (S-FedMV) that takes sequence of multi-view data as input and demonstrated it experimentally. To the best of our knowledge, this framework is the first to consider both vertical and horizontal diversification in the multi-view setting, as well as their sequential federated learning.

Details DOI

YNIMG Journal 2022 Journal Article

Privacy-preserving harmonization via distributed ComBat

  • Andrew A. Chen
  • Chongliang Luo
  • Yong Chen
  • Russell T. Shinohara
  • Haochang Shou

Challenges in clinical data sharing and the need to protect data privacy have led to the development and popularization of methods that do not require directly transferring patient data. In neuroimaging, integration of data across multiple institutions also introduces unwanted biases driven by scanner differences. These scanner effects have been shown by several research groups to severely affect downstream analyses. To facilitate the need of removing scanner effects in a distributed data setting, we introduce distributed ComBat, an adaptation of a popular harmonization method for multivariate data that borrows information across features. We present our fast and simple distributed algorithm and show that it yields equivalent results using data from the Alzheimer's Disease Neuroimaging Initiative. Our method enables harmonization while ensuring maximal privacy protection, thus facilitating a broad range of downstream analyses in functional and structural imaging studies.

Details DOI

TCS Journal 2021 Journal Article

Acyclic edge coloring conjecture is true on planar graphs without intersecting triangles

  • Qiaojun Shu
  • Yong Chen
  • Shuguang Han
  • Guohui Lin
  • Eiji Miyano
  • An Zhang

An acyclic edge coloring of a graph G is a proper edge coloring such that no bichromatic cycles are produced. The acyclic edge coloring conjecture by Fiamčik (1978) and Alon, Sudakov and Zaks (2001) states that every simple graph with maximum degree Δ is acyclically edge ( Δ + 2 ) -colorable. Despite many milestones, the conjecture remains open even for planar graphs. In this paper, we confirm affirmatively the conjecture on planar graphs without intersecting triangles. We do so by first showing, by discharging methods, that every planar graph without intersecting triangles must have at least one of the six specified groups of local structures, and then proving the conjecture by recoloring certain edges in each such local structure and by induction on the number of edges in the graph.

Details DOI

TCS Journal 2021 Journal Article

An improved algorithm for a two-stage production scheduling problem with an outsourcing option

  • Xiaojuan Jiang
  • An Zhang
  • Yong Chen
  • Guangting Chen
  • Kangbok Lee

We consider a two-stage production scheduling problem where each operation can be outsourced or processed in-house. For each operation in the same machine, the ratio of its outsourcing cost to its processing time is constant. The objective is to minimize the sum of the makespan and the total outsourcing cost. It is known that this problem is either polynomial time solvable or NP-hard according to the conditions of the ratios. Even though approximation algorithms for NP-hard cases had been developed, their tight worst-case performance ratios are still open. In this paper, we carefully analyze the approximation algorithms to identify their tight worst-case performance ratios for cases. In one case, we propose a new approximation algorithm with a better and tight worst-case performance ratio. In the process of analyzing the algorithm, we propose a technique utilizing nonlinear optimization.

Details DOI

TCS Journal 2021 Journal Article

Improved hardness and approximation results for single allocation hub location problems

  • Xing Wang
  • Guangting Chen
  • Yong Chen
  • Guohui Lin
  • Yonghao Wang
  • An Zhang

Given a metric graph G = ( V, E, w ) and an integer k, we aim to find a single allocation k-hub location, which is a spanning subgraph consisting of a clique of size k such that every node outside of the clique is adjacent to exactly one node inside the clique. For various objective functions studied in the literature, we present improved hardness and approximation results.

Details DOI

TCS Journal 2020 Journal Article

Approximation algorithms for the three-machine proportionate mixed shop scheduling

  • Longcheng Liu
  • Yong Chen
  • Jianming Dong
  • Randy Goebel
  • Guohui Lin
  • Yue Luo
  • Guanqun Ni
  • Bing Su

A mixed shop is a manufacturing infrastructure designed to process a mixture of a set of flow-shop jobs and a set of open-shop jobs. Mixed shops are in general much more complex to schedule than flow-shops and open-shops, and have been studied since the 1980's. We consider the three machine proportionate mixed shop problem denoted as M 3 | p r p t | C max, in which by “proportionate” each job has equal processing times on all three machines. Koulamas and Kyparisis (2015) [6] showed that the problem is solvable in polynomial time in some very special cases; for the non-solvable case, they proposed a 5/3-approximation algorithm. In this paper, we first present an improved 4/3-approximation algorithm and show that this ratio of 4/3 is asymptotically tight; when the largest job is a flow-shop job, we then present a fully polynomial-time approximation scheme (FPTAS). On the negative side, while the F 3 | p r p t | C max problem is polynomial-time solvable, we show an interesting hardness result that adding one open-shop job to the job set makes the problem NP-hard if this open-shop job is larger than any flow-shop job. We are able to design an FPTAS for this special case too.

Details DOI

YNIMG Journal 2020 Journal Article

High-resolution 3D MR Fingerprinting using parallel imaging and deep learning

  • Yong Chen
  • Zhenghan Fang
  • Sheng-Che Hung
  • Wei-Tang Chang
  • Dinggang Shen
  • Weili Lin

MR Fingerprinting (MRF) is a relatively new imaging framework capable of providing accurate and simultaneous quantification of multiple tissue properties for improved tissue characterization and disease diagnosis. While 2D MRF has been widely available, extending the method to 3D MRF has been an actively pursued area of research as a 3D approach can provide a higher spatial resolution and better tissue characterization with an inherently higher signal-to-noise ratio. However, 3D MRF with a high spatial resolution requires lengthy acquisition times, especially for a large volume, making it impractical for most clinical applications. In this study, a high-resolution 3D MR Fingerprinting technique, combining parallel imaging and deep learning, was developed for rapid and simultaneous quantification of T1 and T2 relaxation times. Parallel imaging was first applied along the partition-encoding direction to reduce the amount of acquired data. An advanced convolutional neural network was then integrated with the MRF framework to extract features from the MRF signal evolution for improved tissue characterization and accelerated mapping. A modified 3D-MRF sequence was also developed in the study to acquire data to train the deep learning model that can be directly applied to prospectively accelerate 3D MRF scans. Our results of quantitative T1 and T2 maps demonstrate that improved tissue characterization can be achieved using the proposed method as compared to prior methods. With the integration of parallel imaging and deep learning techniques, whole-brain (26 × 26 × 18 cm3) quantitative T1 and T2 mapping with 1-mm isotropic resolution were achieved in ~7 min. In addition, a ~7-fold improvement in processing time to extract tissue properties was also accomplished with the deep learning approach as compared to the standard template matching method. All of these improvements make high-resolution whole-brain quantitative MR imaging feasible for clinical applications.

Details DOI

IJCAI Conference 2020 Conference Paper

HyperNews: Simultaneous News Recommendation and Active-Time Prediction via a Double-Task Deep Neural Network

  • Rui Liu
  • Huilin Peng
  • Yong Chen
  • Dell Zhang

Personalized news recommendation can help users stay on top of the current affairs without being overwhelmed by the endless torrents of online news. However, the freshness or timeliness of news has been largely ignored by current news recommendation systems. In this paper, we propose a novel approach dubbed HyperNews which explicitly models the effect of timeliness on news recommendation. Furthermore, we introduce an auxiliary task of predicting the so-called "active-time" that users spend on each news article. Our key finding is that it is beneficial to address the problem of news recommendation together with the related problem of active-time prediction in a multi-task learning framework. Specifically, we train a double-task deep neural network (with a built-in timeliness module) to carry out news recommendation and active-time prediction simultaneously. To the best of our knowledge, such a "kill-two-birds-with-one-stone" solution has seldom been tried in the field of news recommendation before. Our extensive experiments on real-life news datasets have not only confirmed the mutual reinforcement of news recommendation and active-time prediction but also demonstrated significant performance improvements over state-of-the-art news recommendation techniques.

PDF Details DOI

TCS Journal 2020 Journal Article

Open-shop scheduling for unit jobs under precedence constraints

  • Yong Chen
  • Randy Goebel
  • Guohui Lin
  • Bing Su
  • An Zhang

We study open-shop scheduling for unit jobs under precedence constraints, where if one job precedes another job then it has to be finished before the other job can start to be processed. For the three-machine open-shop to minimize the makespan, we first present a simple 5/3-approximation algorithm based on a partition of the job set into agreeable layers using the natural layered representation of the precedence graph, which is directed acyclic. We then show a greedy algorithm to reduce the number of singleton-job layers, resulting in an improved partition, which leads to a 4/3-approximation algorithm. Both approximation algorithms apply to the general m-machine open-shops too.

Details DOI

YNIMG Journal 2019 Journal Article

MR fingerprinting enables quantitative measures of brain tissue relaxation times and myelin water fraction in the first five years of life

  • Yong Chen
  • Meng-Hsiang Chen
  • Kristine R. Baluyot
  • Taylor M. Potts
  • Jordan Jimenez
  • Weili Lin

Quantitative assessments of normative brain development using MRI are of critical importance to gain insights into healthy neurodevelopment. However, quantitative MR imaging poses significant technical challenges and requires prohibitively long acquisition times, making it impractical for pediatric imaging. This is particularly relevant for healthy subjects, where imaging under sedation is not clinically indicated. MR Fingerprinting (MRF), a novel MR imaging framework, provides rapid, efficient, and simultaneous quantification of multiple tissue properties. In this study, a 2D MR Fingerprinting method was developed that achieves a spatial resolution of 1 × 1 × 3 mm3 with rapid and simultaneous quantification of T1, T2 and myelin water fraction (MWF). Phantom experiments demonstrated that accurate measurements of T1 and T2 relaxation times were achieved over a wide range of T1 and T2 values. MRF images were acquired cross-sectionally from 28 typically developing children, 0 to five years old, who were enrolled in the UNC/UMN Baby Connectome Project. Differences associated with age of R1 (=1/T1), R2 (=1/T2) and MWF were obtained from several predefined white matter regions. Both R1 and R2 exhibit a marked increase until ∼20 months of age, followed by a slower increase for all WM regions. In contrast, the MWF remains at a negligible level until ∼6 months of age for all predefined ROIs and gradually increases afterwards. Depending on the brain region, rapid increases are observed between 6 and 12 months to 6–18 months, followed by a slower pace of increase in MWF. Neither relaxivities nor MWF were significantly different between the left and right hemispheres. However, regional differences in age-related R1 and MWF measures were observed across different white matter regions. In conclusion, our results demonstrate that the MRF technique holds great potential for multi-parametric assessments of normative brain development in early childhood.

Details DOI

TCS Journal 2013 Journal Article

Approximation algorithms for two-machine open shop scheduling with batch and delivery coordination

  • Jianming Dong
  • An Zhang
  • Yong Chen
  • Qifan Yang

We consider a scheduling problem with batch and delivery coordination. There are n jobs to be firstly processed by a two-machine open shop at a manufacturing facility, then be delivered to a common customer area by only one vehicle, which is initially located at the facility and has a capacity of c. The objective is to minimize the time when all jobs are completed and delivered to the customer area and the vehicle returns to the facility. For general c, we present a polynomial time approximation algorithm with a worst case ratio of 2. For the case when the vehicle can take only one job in each shipment, we show that there exists a 3 2 -approximation algorithm.

Details DOI

ICRA Conference 2002 Conference Paper

Wireless Drive and Control of a Swimming Microrobot

  • Tao Mei
  • Yong Chen
  • Guoqiang Fu
  • Deyi Kong

A swimming microrobot actuated by two FMP (ferromagnetic polymer) fins was developed. The robot could be driven by external magnetic field wirelessly. Fabrication process and performance of the FMP actuators were presented. Working principle and scaling effect of the robot were analyzed. Experimental results demonstrated that the microrobot could swim under the water, and the speed and direction could be controlled.

Details
v2026.09.13