Hongji Zhu

Mr. Hongji Zhu

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 February 2025 in Mathematics
Verified
Journal Article
Open access
PDF file
LCLiping Chen
Liping Chen
HZHongji Zhu
Hongji Zhu
SHSHUGUANG HAN
SHUGUANG HAN
The node representation learning capability of Graph Convolutional Networks (GCNs) is fundamentally constrained by dynamic instability during feature propagation, yet existing research lacks systematic theoretical analysis of stability control mechanisms. This paper proposes a Stability-Optimized Graph Convolutional Network (SO-GCN) that enhances training stability and feature expressiveness in shallow architectures through continuous–discrete dual-domain stability constraints. By constructing continuous dynamical equations for GCNs and rigorously proving conditional stability under arbitrary parameter dimensions using nonlinear operator theory, we establish theoretical foundations. A Precision Weight Parameter Mechanism is introduced to determine critical Frobenius norm thresholds through feature contraction rates, optimized via differentiable penalty terms. Simultaneously, a Dynamic Step-size Adjustment Mechanism regulates propagation steps based on spectral properties of instantaneous Jacobian matrices and forward Euler discretization. Experimental results demonstrate SO-GCN’s superiority: 1.1–10.7% accuracy improvement on homophilic graphs (Cora/CiteSeer) and 11.22–12.09% enhancement on heterophilic graphs (Texas/Chameleon) compared to conventional GCN. Hilbert–Schmidt Independence Criterion (HSIC) analysis reveals SO-GCN’s superior inter-layer feature independence maintenance across 2–7 layers. This study establishes a novel theoretical paradigm for graph network stability analysis, with practical implications for optimizing shallow architectures in real-world applications.

27 June 2025 in Axioms
Verified
Journal Article
Open access
PDF file
LCLiping Chen
Liping Chen
HZHongji Zhu
Hongji Zhu
SHSHUGUANG HAN
SHUGUANG HAN
Graph attention networks are pivotal for modeling non-Euclidean data, yet they face dual challenges: training oscillations induced by projection-based high-dimensional constraints and gradient anomalies due to poor adaptation to heterophilic structure. To address these issues, we propose LDC-GAT (Lyapunov-Stable Graph Attention Network with Dynamic Filtering and Constraint-Aware Optimization), which jointly optimizes both forward and backward propagation processes. In the forward path, we introduce Dynamic Residual Graph Filtering, which integrates a tunable self-loop coefficient to balance neighborhood aggregation and self-feature retention. This filtering mechanism, constrained by a lower bound on Dirichlet energy, improves multi-head attention via multi-scale fusion and mitigates overfitting. In the backward path, we design the Fro-FWNAdam, a gradient descent algorithm guided by a learning-rate-aware perceptron. An explicit Frobenius norm bound on weights is derived from Lyapunov theory to form the basis of the perceptron. This stability-aware optimizer is embedded within a Frank–Wolfe framework with Nesterov acceleration, yielding a projection-free constrained optimization strategy that stabilizes training dynamics. Experiments on six benchmark datasets show that LDC-GAT outperforms GAT by 10.54% in classification accuracy, which demonstrates strong robustness on heterophilic graphs.

30 June 2026 in Computers
Verified
Journal Article
Open access
PDF file
HZHongji Zhu
Hongji Zhu
LCLiping Chen
Liping Chen
SHSHUGUANG HAN
SHUGUANG HAN
The nesting and cutting of thin materials are critical processes in industrial manufacturing, often involving multi-layer stacking to optimize production efficiency. However, material defects complicate the process, requiring optimization of both layout and defect avoidance under multi-layer heterogeneous constraints. Moreover, existing methods struggle with large search spaces and high computational complexity, limiting their industrial applicability and affecting material utilization and machining accuracy. To address these challenges, we propose HGATrans-MNCD, an intelligent nesting optimization algorithm that integrates defect avoidance and material waste reduction. Initially, areas with high defect overlap are prioritized using an enhanced No-Fit-Polygons strategy to ensure global defect avoidance. A heuristic approach is then employed to optimize the initial nesting sequence. Subsequently, a Transformer-based module leverages prior knowledge to efficiently perturb and refine the sequence, facilitating global optimization. Experiments on a benchmark dataset of multi-layer defect scenarios demonstrate that HGATrans-MNCD effectively addresses irregular defect patterns, enhancing material utilization by 2–8%. Our algorithm performs especially well in scenarios involving spatially coupled defects, offering a novel solution to complex multi-constraint optimization problems.