Author name cluster

Wenjian Wang

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.

10 papers

2 author rows

AAAI Conference 2026 Conference Paper

GCIB: Causal Intervention Guided Graph Information Bottleneck Framework

Hangyuan Du
Rong Wang
Lixin Cui
Gaoxia Jiang
Liang Bai
Wenjian Wang

Graph neural networks (GNNs) have demonstrated impressive performance in a broad spectrum of fields, but always suffer from the generalization problem when confronted with out-of-distribution (OOD) scenarios. Information bottleneck (IB) principle, which endeavors to learn the minimally sufficient representations for downstream tasks, has been shown to be a promising strategy in dealing with this problem. However, the IB-based methods do not inherently distinguish between causal and non-causal parts in the graph, leading to underperforming OOD generalization ability. In this paper, we develop the Graph Causal Information Bottleneck (GCIB) framework, a causal extension of the IB for graph data, which is capable of jointly compressing abundant information and capturing causal dependency from the input graph. Specifically, we endow graph IB with the ability of maintaining causal control by incorporating the underlying causal structure and introducing intervention operation. On this basis, we formulate the learning objective for GCIB and present its specific implementation. Graph representations learned by GCIB can effectively preserve causal information that fundamentally determines graph properties, resulting in outstanding OOD generalization ability. Extensive experiments on both synthetic and real-world datasets demonstrate the superiority of GCIB over state-of-the-art baselines.

PDF Details DOI

ICML Conference 2025 Conference Paper

Can DBNNs Robust to Environmental Noise for Resource-constrained Scenarios?

Wendong Zheng
Junyang Chen
Husheng Guo
Wenjian Wang

Recently, the potential of lightweight models for resource-constrained scenarios has garnered significant attention, particularly in safety-critical tasks such as bio-electrical signal classification and B-ultrasound-assisted diagnostic. These tasks are frequently affected by environmental noise due to patient movement artifacts and inherent device noise, which pose significant challenges for lightweight models (e. g. , deep binary neural networks (DBNNs)) to perform robust inference. A pertinent question arises: can a well-trained DBNN effectively resist environmental noise during inference? In this study, we find that the DBNN’s robustness vulnerability comes from the binary weights and scaling factors. Drawing upon theoretical insights, we propose L1-infinite norm constraints for binary weights and scaling factors, which yield a tighter upper bound compared to existing state-of-the-art (SOTA) methods. Finally, visualization studies show that our approach introduces minimal noise perturbations at the periphery of the feature maps. Our approach outperforms the SOTA method, as validated by several experiments conducted on the bio-electrical and image classification datasets. We hope our findings can raise awareness among researchers about the environmental noise robustness of DBNNs.

Details

NeurIPS Conference 2025 Conference Paper

Noisy Multi-Label Learning through Co-Occurrence-Aware Diffusion

Senyu Hou
Yuru Ren
Gaoxia Jiang
Wenjian Wang

Noisy labels often compel models to overfit, especially in multi-label classification tasks. Existing methods for noisy multi-label learning (NML) primarily follow a discriminative paradigm, which relies on noise transition matrix estimation or small-loss strategies to correct noisy labels. However, they remain substantial optimization difficulties compared to noisy single-label learning. In this paper, we propose a Co-Occurrence-Aware Diffusion (CAD) model, which reformulates NML from a generative perspective. We treat features as conditions and multi-labels as diffusion targets, optimizing the diffusion model for multi-label learning with theoretical guarantees. Benefiting from the diffusion model's strength in capturing multi-object semantics and structured label matrix representation, we can effectively learn the posterior mapping from features to true multi-labels. To mitigate the interference of noisy labels in the forward process, we guide generation using pseudo-clean labels reconstructed from the latent neighborhood space, replacing original point-wise estimates with neighborhood-based proxies. In the reverse process, we further incorporate label co-occurrence constraints to enhance the model's awareness of incorrect generation directions, thereby promoting robust optimization. Extensive experiments on both synthetic (Pascal-VOC, MS-COCO) and real-world (NUS-WIDE) noisy datasets demonstrate that our approach outperforms state-of-the-art methods.

PDF Details

EAAI Journal 2024 Journal Article

A novel Complementary Dual-aware Network for point cloud classification

Rui Hu
Feilong Cao
Wenjian Wang

As an elementary research, three-dimensional (3D) point cloud classification study can further serve high-level downstream applications such as 3D reconstruction, generation, and completion. Recently, excellent performance for synthetic point cloud data classification have achieved, but most of them do not work well on point cloud shape collected from real-world scenarios due to its complexity with noises, varies background, occlusion, etc. As we all known, human can handle it easily. Thus, this paper proposed a Complementary Dual-aware Network (ComDa-Net) inspired by the neurobiological basis of human visual system, aiming to enhance the ability of perceiving 3D objects in real-world scenarios. Specifically, the Essential Perceived Unit (EPU) is proposed to realize the primary complementary dual-aware mechanism through elaborated variational resolutions and receptive fields, then multiple EPUs stack to form the cross-complemented hierarchical system. The proposed method achieves advanced and stable accuracy on the wild-used real-world point cloud benchmarks, and its efficiency in terms of computational and storage is also satisfied, which validates the proposed method’s expected performances. In addition, the proposed method also achieves competitive performance on the synthetic point cloud benchmarks.

Details DOI

JBHI Journal 2024 Journal Article

Dual-Teacher Feature Distillation: A Transfer Learning Method for Insomniac PSG Staging

Lijuan Duan
Yan Zhang
Zhaoyang Huang
Bian Ma
Wenjian Wang
Yuanhua Qiao

Insomnia is the most common sleep disorder linked with adverse long-term medical and psychiatric outcomes. Automatic sleep staging plays a crucial role in aiding doctors to diagnose insomnia disorder. Only a few studies have been conducted to develop automatic sleep staging methods for insomniacs, and most of them have utilized transfer learning methods, which involve pre-training models on healthy individuals and then fine-tuning them on insomniacs. Unfortunately, significant differences in feature distribution between the two subject groups impede the transfer performance, highlighting the need to effectively integrate the features of healthy subjects and insomniacs. In this paper, we propose a dual-teacher cross-domain knowledge transfer method based on the feature-based knowledge distillation to improve the performance of sleep staging for insomniacs. Specifically, the insomnia teacher directly learns from insomniacs and feeds the corresponding domain-specific features into the student network, while the health domain teacher guide the student network to learn domain-generic features. During the training process, we adopt the OFD (Overhaul of Feature Distillation) method to build the health domain teacher. We conducted the experiments to validate the proposed method, using the Sleep-EDF database as the source domain and the CAP-Database as the target domain. The results demonstrate that our method surpasses advanced techniques, achieving an average sleep staging accuracy of 80. 56% on the CAP-Database. Furthermore, our method exhibits promising performance on the private dataset.

Details DOI

AAAI Conference 2024 Conference Paper

Which Is More Effective in Label Noise Cleaning, Correction or Filtering?

Gaoxia Jiang
Jia Zhang
Xuefei Bai
Wenjian Wang
Deyu Meng

Most noise cleaning methods adopt one of the correction and filtering modes to build robust models. However, their effectiveness, applicability, and hyper-parameter insensitivity have not been carefully studied. We compare the two cleaning modes via a rebuilt error bound in noisy environments. At the dataset level, Theorem 5 implies that correction is more effective than filtering when the cleaned datasets have close noise rates. At the sample level, Theorem 6 indicates that confident label noises (large noise probabilities) are more suitable to be corrected, and unconfident noises (medium noise probabilities) should be filtered. Besides, an imperfect hyper-parameter may have fewer negative impacts on filtering than correction. Unlike existing methods with a single cleaning mode, the proposed Fusion cleaning framework of Correction and Filtering (FCF) combines the advantages of different modes to deal with diverse suspicious labels. Experimental results demonstrate that our FCF method can achieve state-of-the-art performance on benchmark datasets.

PDF Details DOI

EAAI Journal 2021 Journal Article

A reliable KNN filling approach for incomplete interval-valued data

Xiaobo Qi
Husheng Guo
Wenjian Wang

Interval-valued data (IVD) is a kind of data where each feature is an interval, and embeds the uncertainty and variability information. However, the missing values (lower or upper bound, or both of them are missed) may occur in the process of data acquisition and transmission, which may lead to obstacles for data processing. To obtain good results, it is important for IVD to process (often ignore or fill) the missing values. A dataset including missing values is named as incomplete interval-valued (IIV) set here. Some ignoring and filling methods for numeric or symbolic data have been proposed, but they cannot be applied for IIV datasets directly. In this work, a reliable k-nearest neighbor approach (RKNN) for incomplete interval-valued data (IIVD) is proposed. A combining rule to determine whether a datum including missing values should be ignored or filled is designed. Those samples with the missing value for each feature will be ignored directly. It is different from existing ignoring methods that need to set the percentage of missing entries. For the rest of missing samples, they will be filled according to their K complete nearest neighbors, which can ensure the filled value more reliable. In so doing, RKNN can exclude a small number of missing samples that may increase uncertainty, and avoid the repetition of the filled values (like median or a fixed constant). The experiment results on 12 synthetic datasets and 4 real-world datasets demonstrate that the proposed method can process the incomplete interval-valued data effectively, and obtain a good classification performance simultaneously.

Details DOI

JMLR Journal 2021 Journal Article

A Unified Sample Selection Framework for Output Noise Filtering: An Error-Bound Perspective

Gaoxia Jiang
Wenjian Wang
Yuhua Qian
Jiye Liang

The existence of output noise will bring difficulties to supervised learning. Noise filtering, aiming to detect and remove polluted samples, is one of the main ways to deal with the noise on outputs. However, most of the filters are heuristic and could not explain the filtering influence on the generalization error (GE) bound. The hyper-parameters in various filters are specified manually or empirically, and they are usually unable to adapt to the data environment. The filter with an improper hyper-parameter may overclean, leading to a weak generalization ability. This paper proposes a unified framework of optimal sample selection (OSS) for the output noise filtering from the perspective of error bound. The covering distance filter (CDF) under the framework is presented to deal with noisy outputs in regression and ordinal classification problems. Firstly, two necessary and sufficient conditions for a fixed goodness of fit in regression are deduced from the perspective of GE bound. They provide the unified theoretical framework for determining the filtering effectiveness and optimizing the size of removed samples. The optimal sample size has the adaptability to the environmental changes in the sample size, the noise ratio, and noise variance. It offers a choice of tuning the hyper-parameter and could prevent filters from overcleansing. Meanwhile, the OSS framework can be integrated with any noise estimator and produces a new filter. Then the covering interval is proposed to separate low-noise and high-noise samples, and the effectiveness is proved in regression. The covering distance is introduced as an unbiased estimator of high noises. Further, the CDF algorithm is designed by integrating the cover distance with the OSS framework. Finally, it is verified that the CDF not only recognizes noise labels correctly but also brings down the prediction errors on real apparent age data set. Experimental results on benchmark regression and ordinal classification data sets demonstrate that the CDF outperforms the state-of-the-art filters in terms of prediction ability, noise recognition, and efficiency. [abs] [ pdf ][ bib ] &copy JMLR 2021. ( edit, beta )

PDF Details

EAAI Journal 2020 Journal Article

An accelerator for online SVM based on the fixed-size KKT window

Husheng Guo
Aijuan Zhang
Wenjian Wang

Support vector machine (SVM), as a general and useful supervised learning tool, is facing with some challenges such as low learning efficiency, poor generalization performance, noise sensitivity, etc. when it is applied to online learning tasks. To overcome these limitations, an accelerator model based on window technology and the KKT conditions for online SVM learning is proposed in this paper. The proposed model is not an independent online algorithm but may be regarded as an accelerator for other online SVM learning algorithms, and it constructs working set of SVM by a fixed-size window with the samples which violate the KKT conditions. The relationship between Lagrangain multipliers in dual problem of SVM and KKT conditions are analyzed in the case of online learning. On this basis, a fixed-size KKT window can be constructed according to whether the samples violate KKT conditions or not. Then, it takes the samples that violate the KKT conditions as the training window, which not only makes the training samples with the same size each time, but also ensures that all samples are useful for the hyperplane updating (it means that the classifier can be updated more smoothly). Two typical and specific online SVM algorithms are used as baseline, and the corresponding speeding online SVM learning algorithms with ”X+accelerator” models are proposed to testing the performance of the proposed accelerator. Comprehensive experiments clearly show that the proposed model can accelerate the online learning process effectively and has good robustness and generalization performance.

Details DOI

IJCAI Conference 2015 Conference Paper

Robust Multiple Kernel K-means Using L21-Norm

Liang Du
Peng Zhou
Lei Shi
Hanmo Wang
Mingyu Fan
Wenjian Wang
Yi-Dong Shen

The k-means algorithm is one of the most often used method for data clustering. However, the standard k-means can only be applied in the original feature space. The kernel k-means, which extends k-means into the kernel space, can be used to capture the non-linear structure and identify arbitrarily shaped clusters. Since both the standard k-means and kernel k-means apply the squared error to measure the distances between data points and cluster centers, a few outliers will cause large errors and dominate the objection function. Besides, the performance of kernel method is largely determined by the choice of kernel. Unfortunately, the most suitable kernel for a particular task is often unknown in advance. In this paper, we first present a robust kmeans using `2, 1-norm in the feature space and then extend it to the kernel space. To recap the powerfulness of kernel methods, we further propose a novel robust multiple kernel k-means (RMKKM) algorithm that simultaneously finds the best clustering label, the cluster membership and the optimal combination of multiple kernels. An alternating iterative schema is developed to find the optimal value. Extensive experiments well demonstrate the effectiveness of the proposed algorithms.

PDF Details