Research map / Graph learning / GNN architectures
DepthLGP: Learning Embeddings of Out-of-Sample Nodes in Dynamic Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add neural transformation g(.) on top of hLGP (DepthLGP vs hLGP) (ablation) | DepthLGP · DBLP, node2vec | Macro-F1 (%): 60.89 -> 62.63 | Helped |
| Add neural transformation g(.) (ablation) | DepthLGP · BlogCatalog, GraRep | Macro-F1 (%): 15.94 -> 18.45 | Helped |
| Add neural transformation g(.) (ablation) | DepthLGP · PPI, LINE | Micro-F1 (%): 20.71 -> 21.42 | Helped |
| Add neural transformation g(.) (ablation) | DepthLGP · PPI, node2vec | AUC (%): 53.16 -> 55.37 | Helped |
| Use high-order Laplacian GP (hLGP) instead of LabelProp (ablation) | hLGP · DBLP, node2vec | Macro-F1 (%): 59.25 -> 60.89 | Helped |
| Use DepthLGP instead of LabelProp (ablation) | DepthLGP · PPI, node2vec | AUC (%): 46.79 -> 55.37 | Helped |
| Use DepthLGP instead of LocalAvg (ablation) | DepthLGP · BlogCatalog, node2vec | Micro-F1 (%): 34.83 -> 39.64 | Helped |
| Use DepthLGP instead of LabelProp (ablation) | DepthLGP · BlogCatalog, LINE | AUC (%): 54.71 -> 55.53 | No clear effect |
| Use DepthLGP out-of-sample inference vs rerunning embedding on evolved network (upper bound) (ablation) | DepthLGP · DBLP, node2vec | Macro-F1 (%): 64.87 -> 62.63 | Hurt |
| Integrate DepthLGP into node2vec training (node2vec++) (ablation) | node2vec · DBLP out-of-sample | AUC (%): 82.54 -> 85.88 | Helped |
| Integrate DepthLGP into node2vec (node2vec++) (ablation) | node2vec · DBLP in-sample | AUC (%): 90.02 -> 89.20 | Hurt |
| Fix second-order proximity coefficient zeta to zero (ablation) | DepthLGP · theory | expressive power | Hurt |
Closest papers by meaning
- Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking (ICLR 2018 · similarity 0.90)
- Network Structure and Transfer Behaviors Embedding via Deep Prediction Model (AAAI 2019 · similarity 0.89)
- DeepWalking Backwards: From Embeddings Back to Graphs (ICML 2021 · similarity 0.88)
- Dynamic Embedding on Textual Networks via a Gaussian Process (AAAI 2020 · similarity 0.88)
- How Low Can You Go? Searching for the Intrinsic Dimensionality of Complex Networks using Metric Node Embeddings (ICLR 2025 · similarity 0.88)
- Asymptotics of $\ell_2$ Regularized Network Embeddings (NeurIPS 2022 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.