YHYunzhi Hao

Yunzhi Hao

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yunzhi Hao has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Similarity-Guided Graph Learning with Embedding-Space Oversampling for Anomaly Detection in Heterophilic Cryptocurrency Transaction Networks

    by Jing Cao, Da He, Yunzhi Hao

    The rapid development of cryptocurrency has led to an urgent demand for anomaly detection targeting illegal transaction activities. Cryptocurrency transaction networks have two core characteristics: class imbalance and connection heterophily. Existing anomaly-detection methods tend to cause over-smoothing of anomaly nodes’ features and have low recall rates for anomaly nodes, resulting in poor detection performance. Although some studies have noticed these challenges separately, there is still a lack of a unified solution framework. This paper proposes a cryptocurrency anomaly-detection framework specially designed for cryptocurrency transaction scenarios. It consists of three core modules: Heterophily-adaptive Aggregation Module, Generation-based Minority Class Oversampling Module, and Ensemble Anomaly-Detection Classifier. The Heterophily-adaptive Aggregation Module enables the framework to resist heterophily noise during learning node embeddings by adaptively adjusting the aggregation weights of neighboring nodes. The Generation-based Minority Class Oversampling Module augments the learned minority node embeddings in the embedding space to obtain a more balanced class distribution. Finally, the Ensemble Anomaly-Detection Classifier is used to predict the detection result. Experimental results on multiple cryptocurrency transaction datasets show that the proposed framework significantly outperforms traditional machine learning methods and mainstream graph neural network models.

  • Open Access

    A Physics-Constrained Dual-Stream Dynamic Framework for Wind Power Forecasting Under Extreme Weather

    by Yunzhi Hao, Jing Cao

    Accurate wind power forecasting is essential for ensuring power grid stability and facilitating the large-scale integration of renewable energy, yet it faces significant challenges due to the randomness, variability, and intermittency of wind resources and the increasing frequency of extreme weather events. Existing data-driven approaches often struggle to balance temporal continuity with meteorological sensitivity, leading to lag effects during rapid fluctuations, and frequently generate predictions that violate physical domain knowledge. To address these limitations, this paper proposes a dual-stream architecture to decouple temporal dependencies and spatial–meteorological mappings, utilizing a Physics-Informed GRU (PI-GRU) and an Enhanced Random Forest (ERF). Both streams are strictly bounded by physical constraints. Furthermore, a scenario-aware adaptive fusion mechanism is introduced to dynamically adjust the model’s reliance on each stream based on real-time wind speed gradients and volatility indices. Extensive experiments were conducted using a comprehensive dataset from three coastal wind farms over 8 months, encompassing stable regimes and extreme weather events. Evaluating across both 1-day and 4-day forecast horizons, the results demonstrate that our method significantly outperforms state-of-the-art baselines, proving its robustness and practical value for grid security and dispatch optimization.


  • Spatiotemporal-Augmented Graph Neural Networks for Human Mobility Simulation

    by Yu Wang, Tongya Zheng, Shunyu Liu, Zunlei Feng, Kaixuan Chen, Yunzhi Hao, Mingli Song

    Human mobility patterns have shown significant applications in policy-decision scenarios and economic behavior researches. The human mobility simulation task aims to generate human mobility trajectories given a small set of trajectory data, which have aroused much concern due to the scarcity and sparsity of human mobility data. Existing methods mostly rely on the static relationships of locations, while largely neglect the dynamic spatiotemporal effects of locations. On the one hand, spatiotemporal correspondences of visit distributions reveal the spatial proximity and the functionality similarity of locations. On the other hand, the varying durations in different locations hinder the iterative generation process of the mobility trajectory. Therefore, we propose a novel framework to model the dynamic spatiotemporal effects of locations, namely S patio T emporal- A ugmented g R aph neural networks (STAR). The STAR framework designs various spatiotemporal graphs to capture the spatiotemporal correspondences and builds a novel dwell branch to simulate the varying durations in locations, which is finally optimized in an adversarial manner. The comprehensive experiments over four real datasets for the human mobility simulation have verified the superiority of STAR to state-of-the-art methods.


  • Heterogeneous Graph Prototypical Networks for Few-Shot Node Classification

    by Yunzhi Hao, Mengfan Wang, Xingen Wang, Tongya Zheng, Xinyu Wang, Wenqi Huang, Chun Chen


  • Attribution Guided Layerwise Knowledge Amalgamation from Graph Neural Networks

    by Yunzhi Hao, Yu Wang, Shunyu Liu, Tongya Zheng, Xingen Wang, Xinyu Wang, Mingli Song, Wenqi Huang, Chun Chen

  • Temporal Aggregation and Propagation Graph Neural Networks for Dynamic Representation

    by Tongya Zheng, Xinchao Wang, Zunlei Feng, Jie Song, Yunzhi Hao, Mingli Song, Xingen Wang, Xinyu Wang, Chun Chen

    Temporal graphs exhibit dynamic interactions between nodes over continuous time, whose topologies evolve with time elapsing. The whole temporal neighborhood of nodes reveals the varying preferences of nodes. However, previous works usually generate dynamic representation with limited neighbors for simplicity, which results in both inferior performance and high latency of online inference. Therefore, in this paper, we propose a novel method of temporal graph convolution with the whole neighborhood, namely Temporal Aggregation and Propagation Graph Neural Networks (TAP-GNN). Specifically, we firstly analyze the computational complexity of the dynamic representation problem by unfolding the temporal graph in a message-passing paradigm. The expensive complexity motivates us to design the AP (aggregation and propagation) block, which significantly reduces the repeated computation of historical neighbors. The final TAP-GNN supports online inference in the graph stream scenario, which incorporates the temporal information into node embeddings with a temporal activation function and a projection layer besides several AP blocks. Experimental results on various real-life temporal networks show that our proposed TAP-GNN outperforms existing temporal graph methods by a large margin in terms of both predictive performance and online inference latency. Our code is available at https://github.com/doujiang-zheng/TAP-GNN .


  • Transition Propagation Graph Neural Networks for Temporal Networks

    by Tongya Zheng, Zunlei Feng, Tianli Zhang, Yunzhi Hao, Mingli Song, Xingen Wang, Xinyu Wang, Ji Zhao, Chun Chen

    Researchers of temporal networks (e.g., social networks and transaction networks) have been interested in mining dynamic patterns of nodes from their diverse interactions. Inspired by recently powerful graph mining methods like skip-gram models and graph neural networks (GNNs), existing approaches focus on generating temporal node embeddings sequentially with nodes’ sequential interactions. However, the sequential modeling of previous approaches cannot handles the transition structure between nodes’ neighbors with limited memorization capacity. In detail, an effective method for the transition structures is required to both model nodes’ personalized patterns adaptively and capture node dynamics accordingly. In this article, we propose a method, namely transition propagation graph neural networks (TIP-GNN), to tackle the challenges of encoding nodes’ transition structures. The proposed TIP-GNN focuses on the bilevel graph structure in temporal networks: besides the explicit interaction graph, a node’s sequential interactions can also be constructed as a transition graph. Based on the bilevel graph, TIP-GNN further encodes transition structures by multistep transition propagation and distills information from neighborhoods by a bilevel graph convolution. Experimental results over various temporal networks reveal the efficiency of our TIP-GNN, with at most 7.2% improvements of accuracy on temporal link prediction. Extensive ablation studies further verify the effectiveness and limitations of the transition propagation module. Our code is available at https://github.com/doujiang-zheng/TIP-GNN.


  • Walking With Attention: Self-Guided Walking for Heterogeneous Graph Embedding

    by Yunzhi Hao, Xinchao Wang, Xingen Wang, Xinyu Wang, Chun Chen, Mingli Song

    Heterogeneous graph embedding aims at learning low-dimensional representations from a graph featuring nodes and edges of diverse natures, and meanwhile preserving the underlying topology. Existing approaches along this line have largely relied on meta-paths , which are by nature hand-crafted and pre-defined transition rules, so as to explore the semantics of a graph. Despite the promising results, defining meta-paths requires domain knowledge, and thus when the test distribution deviates from the priors, such methods are prone to errors. In this paper, we propose a self-learning scheme for heterogeneous graph embedding, termed as self-guided walk (SILK), that bypasses meta-paths and learns adaptive attentions for node walking. SILK assumes no prior knowledge or annotation is provided, and conducts a customized random walk to encode the contexts of the heterogeneous graph of interest. Specifically, this is achieved via maintaining a dynamically-updated guidance matrix that records the node-conditioned transition potentials. Experimental results on four real-world datasets demonstrate that SILK significantly outperforms state-of-the-art methods.