arXiv subjects

Jianshe Wu

Publications and source records attributed to Jianshe Wu.

3 recordsLinked to original sources

arXiv2025

HyperEvent: A Strong Baseline for Dynamic Link Prediction via Relative Structural Encoding

Learning representations for continuous-time dynamic graphs is critical for dynamic link prediction. While recent methods have become increasingly complex, the field lacks a strong and informative baseline to reliably gauge progress. This paper proposes HyperEvent, a simple approach that captures relative structural patterns in event sequences through an intuitive encoding mechanism. As a straightforward baseline, HyperEvent leverages relative structural encoding to identify meaningful event sequences without complex parameterization. By combining these interpretable features with a lightweight transformer classifier, HyperEvent reframes link prediction as event structure recognition. Despite its simplicity, HyperEvent achieves competitive results across multiple benchmarks, often matching the performance of more complex models. This work demonstrates that effective modeling can be achieved through simple structural encoding, providing a clear reference point for evaluating future advancements.

cs.LG↗

arXiv2025

Self-similarity Analysis in Deep Neural Networks

Current research has found that some deep neural networks exhibit strong hierarchical self-similarity in feature representation or parameter distribution. However, aside from preliminary studies on how the power-law distribution of weights across different training stages affects model performance,there has been no quantitative analysis on how the self-similarity of hidden space geometry influences model weight optimization, nor is there a clear understanding of the dynamic behavior of internal neurons. Therefore, this paper proposes a complex network modeling method based on the output features of hidden-layer neurons to investigate the self-similarity of feature networks constructed at different hidden layers, and analyzes how adjusting the degree of self-similarity in feature networks can enhance the classification performance of deep neural networks. Validated on three types of networks MLP architectures, convolutional networks, and attention architectures this study reveals that the degree of self-similarity exhibited by feature networks varies across different model architectures. Furthermore, embedding constraints on the self-similarity of feature networks during the training process can improve the performance of self-similar deep neural networks (MLP architectures and attention architectures) by up to 6 percentage points.

cs.LG↗

arXiv2022

Budgeted Influence Maximization via Boost Simulated Annealing in Social Networks

Due to much closer to real application scenarios,the budgeted influence maximization (BIM) problem has attracted great attention among researchers. As a variant of the influence maximization (IM) problem, the BIM problem aims at mining several nodes with different costs as seeds with limited budget to maximize the influence as possible. By first activating these seed nodes and spreading influence under the given propagation model, the maximized spread of influence can be reached in the network. Several approaches have been proposed for BIM. Most of them are modified versions of the greedy algorithm, which work well on the IM but seems inefficient for the BIM because huge time consuming is inevitable. Recently, some intelligence algorithms are proposed in order to reduce the running time, but analysis shows that they cannot fully utilize the relationships between nodes in networks, which will result in influence loss. Inspired by this, we propose an efficient method based on boosted simulated annealing (SA) algorithm in this paper. Three heuristic strategies are proposed to improve the performance and speed up the proposed algorithm. Experimental results on both real world and synthetic networks demonstrate that the proposed boosted SA performs much better than existed algorithms on performance with almost equal or less running time.

cs.SI↗