Community-Aware Variational Autoencoder for Continuous Dynamic Networks
Authors: Junwei Cheng, Chaobo He, Pengxing Feng, Weixiong Liu, Kunlin Han, Yong Tang
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results demonstrate that the proposed CT-VAE and CTCAVAE achieve more favorable performance compared with the state-of-the-art baselines. ... Extensive experiments on six real-world datasets verify that the proposed CT-VAE outperforms state-of-the-art baselines and demonstrate the promising performance of CT-CAVAE in specific data scenarios. |
| Researcher Affiliation | Collaboration | 1School of Computer Science, South China Normal University 2Department of Electrical Engineering, City University of Hong Kong 3CMT US Holdings LLC 4Computer Science Department, University of Southern California |
| Pseudocode | No | The paper describes methods and processes in paragraph text and mathematical formulations but does not contain any clearly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code, nor does it provide links to any code repositories. |
| Open Datasets | Yes | To comprehensively evaluate our methods, several continuous dynamic networks from various real-world domains are used. Specifically, these include a co-author network (DBLP), two citation networks (ar Xiv CS and ar Xiv AI), a brain network (Brain), a patent citation network (Patent), and a high school student interaction network (School). ... For a detailed introduction to these datasets, refer to (Liu et al. 2024b). |
| Dataset Splits | No | The paper mentions using six real-world datasets but does not provide specific details regarding how these datasets were split into training, validation, or test sets for reproduction. |
| Hardware Specification | Yes | All trials have been conducted on Intel Core i76700 CPUs and NVIDIA RTX 3090 GPUs. |
| Software Dependencies | No | The paper mentions that 'Adam is employed for training' but does not specify any software libraries or their version numbers, such as Python, PyTorch, TensorFlow, or specific Adam optimizer versions. |
| Experiment Setup | Yes | For CT-VAE and CTCAVAE, the Adam is employed for training, with the learning rate selected from {1e 2, 1e 3, 2e 4, 5e 5}. Additionally, we set the negative sampling size to 2 and the history window to 1 in the Hawkes process. To ensure the fairness of our experiments, we initialize the representation Z using node2vec for all methods that require initialization. Furthermore, our proposed methods are trained for 100 epochs with a batch size of 1024. |