Search Results for author: Zengfeng Huang

Found 33 papers, 16 papers with code

基于多质心异质图学习的社交网络用户建模(User Representation Learning based on Multi-centroid Heterogeneous Graph Neural Networks)

no code implementations CCL 2021 Shangyi Ning, Guanying Li, Qin Chen, Zengfeng Huang, Baohua Zhou, Zhongyu Wei

“用户建模已经引起了学术界和工业界的广泛关注。现有的方法大多侧重于将用户间的人际关系融入社区, 而用户生成的内容(如帖子)却没有得到很好的研究。在本文中, 我们通过实际舆情传播相关的分析表明, 在舆情传播过程中对用户属性进行研究的重要作用, 并且提出了用户资料数据的筛选方法。同时, 我们提出了一种通过异构多质心图池为用户捕获更多不同社区特征的建模。我们首先构造了一个由用户和关键字组成的异质图, 并在其上采用了一个异质图神经网络。为了方便用户建模的图表示, 提出了一种多质心图池化机制, 将多质心的集群特征融入到表示学习中。在三个基准数据集上的大量实验表明了该方法的有效性。”

Representation Learning

Optimal Matrix Sketching over Sliding Windows

no code implementations13 May 2024 Hanyan Yin, Dongxie Wen, Jiajun Li, Zhewei Wei, Xiao Zhang, Zengfeng Huang, Feifei Li

Matrix sketching, aimed at approximating a matrix $\boldsymbol{A} \in \mathbb{R}^{N\times d}$ consisting of vector streams of length $N$ with a smaller sketching matrix $\boldsymbol{B} \in \mathbb{R}^{\ell\times d}, \ell \ll N$, has garnered increasing attention in fields such as large-scale data analytics and machine learning.

StructComp: Substituting Propagation with Structural Compression in Training Graph Contrastive Learning

1 code implementation8 Dec 2023 Shengzhong Zhang, Wenjie Yang, Xinyuan Cao, Hongwei Zhang, Zengfeng Huang

This allows the encoder not to perform any message passing during the training stage, and significantly reduces the number of sample pairs in the contrastive loss.

Contrastive Learning

Understanding Community Bias Amplification in Graph Representation Learning

no code implementations8 Dec 2023 Shengzhong Zhang, Wenjie Yang, Yimin Zhang, Hongwei Zhang, Divin Yan, Zengfeng Huang

In this work, we discover a phenomenon of community bias amplification in graph representation learning, which refers to the exacerbation of performance bias between different classes by graph representation learning.

Contrastive Learning Data Augmentation +1

FreshGNN: Reducing Memory Access via Stable Historical Embeddings for Graph Neural Network Training

no code implementations18 Jan 2023 Kezhao Huang, Haitian Jiang, Minjie Wang, Guangxuan Xiao, David Wipf, Xiang Song, Quan Gan, Zengfeng Huang, Jidong Zhai, Zheng Zhang

A key performance bottleneck when training graph neural network (GNN) models on large, real-world graphs is loading node features onto a GPU.

ASGNN: Graph Neural Networks with Adaptive Structure

no code implementations3 Oct 2022 Zepeng Zhang, Songtao Lu, Zengfeng Huang, Ziping Zhao

In this work, we propose a novel interpretable message passing scheme with adaptive structure (ASMP) to defend against adversarial attacks on graph structure.

Node Classification

Optimal Clustering with Noisy Queries via Multi-Armed Bandit

no code implementations12 Jul 2022 Jinghui Xia, Zengfeng Huang

In particular, a new polynomial time algorithm with $O(\frac{n(k+\log n)}{\delta^2} + \text{poly}(k,\frac{1}{\delta}, \log n))$ queries is proposed.


Transformers from an Optimization Perspective

1 code implementation27 May 2022 Yongyi Yang, Zengfeng Huang, David Wipf

Deep learning models such as the Transformer are often constructed by heuristics and experience.

BSAL: A Framework of Bi-component Structure and Attribute Learning for Link Prediction

1 code implementation18 Apr 2022 Bisheng Li, Min Zhou, Shengzhong Zhang, Menglin Yang, Defu Lian, Zengfeng Huang

Regarding missing link inference of diverse networks, we revisit the link prediction techniques and identify the importance of both the structural and attribute information.

Attribute Graph Classification +2

Exploiting Neighbor Effect: Conv-Agnostic GNNs Framework for Graphs with Heterophily

1 code implementation19 Mar 2022 Jie Chen, Shouzhen Chen, Junbin Gao, Zengfeng Huang, Junping Zhang, Jian Pu

Moreover, we propose a simple yet effective Conv-Agnostic GNN framework (CAGNNs) to enhance the performance of most GNNs on heterophily datasets by learning the neighbor effect for each node.

Node Classification

Implicit vs Unfolded Graph Neural Networks

no code implementations12 Nov 2021 Yongyi Yang, Tang Liu, Yangkun Wang, Zengfeng Huang, David Wipf

It has been observed that graph neural networks (GNN) sometimes struggle to maintain a healthy balance between the efficient modeling long-range dependencies across nodes while avoiding unintended consequences such oversmoothed node representations or sensitivity to spurious edges.

Graph Attention Node Classification

Lipschitz Bandits with Batched Feedback

no code implementations19 Oct 2021 Yasong Feng, Zengfeng Huang, Tianyu Wang

Specifically, we show that for a $T$-step problem with Lipschitz reward of zooming dimension $d_z$, our algorithm achieves theoretically optimal (up to logarithmic factors) regret rate $\widetilde{\mathcal{O}}\left(T^{\frac{d_z+1}{d_z+2}}\right)$ using only $ \mathcal{O} \left( \log\log T\right) $ batches.

Why Propagate Alone? Parallel Use of Labels and Features on Graphs

no code implementations ICLR 2022 Yangkun Wang, Jiarui Jin, Weinan Zhang, Yongyi Yang, Jiuhai Chen, Quan Gan, Yong Yu, Zheng Zhang, Zengfeng Huang, David Wipf

In this regard, it has recently been proposed to use a randomly-selected portion of the training labels as GNN inputs, concatenated with the original node features for making predictions on the remaining labels.

Node Property Prediction Property Prediction

One-Bit Matrix Completion with Differential Privacy

no code implementations2 Oct 2021 Zhengpin Li, Zheng Wei, Zengfeng Huang, Xiaojun Mao, Jian Wang

In this paper, we propose a unified framework for ensuring a strong privacy guarantee of one-bit matrix completion with DP.

Collaborative Filtering Matrix Completion +2

Stabilized Self-training with Negative Sampling on Few-labeled Graph Data

no code implementations29 Sep 2021 Ziang Zhou, Jieming Shi, Shengzhong Zhang, Zengfeng Huang, Qing Li

Therefore, we propose an effective framework, Stabilized self-training with Negative sampling (SN), which is applicable to existing GNNs to stabilize the training process and enhance the training data, and consequently, boost classification accuracy on graphs with few labeled data.

Benchmarking Node Classification

Discrete-time Temporal Network Embedding via Implicit Hierarchical Learning in Hyperbolic Space

1 code implementation8 Jul 2021 Menglin Yang, Min Zhou, Marcus Kalander, Zengfeng Huang, Irwin King

To explore these properties of a complex temporal network, we propose a hyperbolic temporal graph network (HTGN) that fully takes advantage of the exponential capacity and hierarchical awareness of hyperbolic geometry.

Graph Embedding Link Prediction +1

Learning Based Proximity Matrix Factorization for Node Embedding

1 code implementation10 Jun 2021 Xingyi Zhang, Kun Xie, Sibo Wang, Zengfeng Huang

Recent progress on node embedding shows that proximity matrix factorization methods gain superb performance and scale to large graphs with millions of nodes.

Link Prediction Node Classification

Scaling Up Graph Neural Networks Via Graph Coarsening

1 code implementation9 Jun 2021 Zengfeng Huang, Shengzhong Zhang, Chong Xi, Tang Liu, Min Zhou

Scalability of graph neural networks remains one of the major challenges in graph machine learning.

Stochastic Optimization

Understanding Bandits with Graph Feedback

no code implementations NeurIPS 2021 Houshuang Chen, Zengfeng Huang, Shuai Li, Chihao Zhang

We propose the notions of the fractional weak domination number $\delta^*$ and the $k$-packing independence number capturing upper bound and lower bound for the regret respectively.

Graph Neural Networks Inspired by Classical Iterative Algorithms

1 code implementation10 Mar 2021 Yongyi Yang, Tang Liu, Yangkun Wang, Jinjing Zhou, Quan Gan, Zhewei Wei, Zheng Zhang, Zengfeng Huang, David Wipf

Despite the recent success of graph neural networks (GNN), common architectures often exhibit significant limitations, including sensitivity to oversmoothing, long-range dependencies, and spurious edges, e. g., as can occur as a result of graph heterophily or adversarial attacks.

Node Classification

Simple and Deep Graph Convolutional Networks

4 code implementations ICML 2020 Ming Chen, Zhewei Wei, Zengfeng Huang, Bolin Ding, Yaliang Li

We propose the GCNII, an extension of the vanilla GCN model with two simple yet effective techniques: {\em Initial residual} and {\em Identity mapping}.

Graph Classification Graph Regression +3

SCE: Scalable Network Embedding from Sparsest Cut

1 code implementation30 Jun 2020 Shengzhong Zhang, Zengfeng Huang, Haicang Zhou, Ziang Zhou

A key of success to such contrastive learning methods is how to draw positive and negative samples.

Contrastive Learning Network Embedding

Optimal Sparsity-Sensitive Bounds for Distributed Mean Estimation

1 code implementation NeurIPS 2019 Zengfeng Huang, Ziyue Huang, Yilei Wang, Ke Yi

We consider the problem of estimating the mean of a set of vectors, which are stored in a distributed system.

Higher-order Weighted Graph Convolutional Networks

no code implementations11 Nov 2019 Songtao Liu, Lingwei Chen, Hanze Dong, ZiHao Wang, Dinghao Wu, Zengfeng Huang

Graph Convolution Network (GCN) has been recognized as one of the most effective graph models for semi-supervised learning, but it extracts merely the first-order or few-order neighborhood information through information propagation, which suffers performance drop-off for deeper structure.

Node Classification

Effective Stabilized Self-Training on Few-Labeled Graph Data

1 code implementation7 Oct 2019 Ziang Zhou, Jieming Shi, Shengzhong Zhang, Zengfeng Huang, Qing Li

However, under extreme cases when very few labels are available (e. g., 1 labeled node per class), GNNs suffer from severe performance degradation.

Benchmarking Model Selection +1

Few-shot Classification on Graphs with Structural Regularized GCNs

no code implementations ICLR 2019 Shengzhong Zhang, Ziang Zhou, Zengfeng Huang, Zhongyu Wei

We consider the fundamental problem of semi-supervised node classification in attributed graphs with a focus on \emph{few-shot} learning.

Classification Few-Shot Learning +2

Near Optimal Frequent Directions for Sketching Dense and Sparse Matrices

no code implementations ICML 2018 Zengfeng Huang

In particular, we provide new space-optimal algorithms with faster running times.

Cannot find the paper you are looking for? You can Submit a new open access paper.