Research map / Graph learning / Knowledge and temporal graphs
Incorporating GAN for Negative Sampling in Knowledge Representation Learning
Peifeng Wang, Shuangyin Li, Rong Pan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Use GAN-generated negative triplets instead of random negatives with TransD | TransD · FB15K | Hits@10 (%): 77.3 -> 77.6 | Improved |
| Use GAN-generated negative triplets instead of random negatives with TransH | TransH · WN18 | Hits@10 (%): 86.7 -> 94.0 | Improved |
| Use GAN-generated negative triplets instead of random negatives with TransE + A-LSTM | TransE + A-LSTM · WN18 | Mean Rank: 123 -> 114 | Improved |
| Use GAN-generated negative triplets instead of random negatives with TransD | TransD · FB13 | Accuracy (%): 89.1 -> 89.7 | Improved |
| Use GAN-generated negative triplets instead of random negatives with TransR | TransR · WN11 | Accuracy (%): 85.9 -> 86.3 | Hurt |
| Initialize the discriminator from a model pretrained with random negative sampling rather than train from scratch (ablation) | GAN-based knowledge embedding · FB15K | Hits@10 (%): TransD: 74.0 -> 77.6 | Mixed |
| Initialize the discriminator from a model pretrained with random negative sampling rather than train from scratch (ablation) | GAN-based knowledge embedding · FB13 | Accuracy (%): TransD: 85.3 -> 89.7 | Mixed |
| Use GAN negative sampling with TransR | TransR · FB15K | Hits@10 (%): 68.7 -> 58.3 | Hurt |
| Use GAN negative sampling with TransD | TransD · FB15K | Hits@10 (%): 77.3 -> 74.0 | Hurt |
Closest papers by meaning
- Generative Adversarial Zero-Shot Relational Learning for Knowledge Graphs (AAAI 2020 · similarity 0.88)
- RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space (ICLR 2019 · similarity 0.86)
- Nickel and Diming Your GAN: A Dual-Method Approach to Enhancing GAN Efficiency via Knowledge Distillation (ECCV 2024 · similarity 0.86)
- Comprehensive Analysis of Negative Sampling in Knowledge Graph Representation Learning (ICML 2022 · similarity 0.85)
- Semantic Relation Preserving Knowledge Distillation for Image-to-Image Translation (ECCV 2020 · similarity 0.85)
- Representation Learning of Knowledge Graphs with Entity Descriptions (AAAI 2016 · similarity 0.85)
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.
