Summary of Ctrl: Continuous-time Representation Learning on Temporal Heterogeneous Information Network, by Chenglin Li and Yuanzhen Xie and Chenyun Yu and Lei Cheng and Bo Hu and Zang Li and Di Niu
CTRL: Continuous-Time Representation Learning on Temporal Heterogeneous Information Network
by Chenglin Li, Yuanzhen Xie, Chenyun Yu, Lei Cheng, Bo Hu, Zang Li, Di Niu
First submitted to arxiv on: 11 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Continuous-Time Representation Learning (CTRL) model tackles the challenges of scalable deep learning on time-varying heterogeneous information networks (HINs). Specifically, it handles new nodes or edges and captures the evolution of high-order topological structures. The CTRL model integrates three components: heterogeneous attention, edge-based Hawkes process, and dynamic centrality. It is trained for future event prediction, outperforming state-of-the-art approaches on three benchmark datasets. Ablation studies demonstrate the effectiveness of the model design. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to learn representations on temporal graphs, which are important for many real-world applications like citation networks. The old methods can’t handle new nodes or edges, and they don’t understand how the graph changes over time. The new model, called CTRL, is better because it considers both node features and temporal structures. It’s trained to predict future events on a graph, which helps it learn about the changing relationships between nodes. The results show that this approach works well and beats other methods. |
Keywords
» Artificial intelligence » Attention » Deep learning » Representation learning
