AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Knowledge Graph Completion via Complex Tensor Factorization

Théo Trouillon, Christopher R. Dance, Éric Gaussier, Johannes Welbl, Sebastian Riedel, Guillaume Bouchard

JMLR 2017 · Graph learning › Knowledge and temporal graphs · approach: Knowledge graph embeddings · 214 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use complex-valued embeddings (ComplEx) instead of real DistMultknowledge graph embedding · WN18filtered MRR: 0.822 -> 0.941Improved
Use complex-valued embeddings (ComplEx) instead of real DistMultknowledge graph embedding · FB15Kfiltered MRR: 0.654 -> 0.692Improved
Use ComplEx with log-likelihood loss instead of HolE with max-margin lossknowledge graph embedding · FB15Kfiltered MRR: 0.524 -> 0.692Improved
Use shared entity embeddings (ComplEx) instead of separate subject/object embeddings (CP)knowledge graph embedding · WN18filtered MRR: 0.075 -> 0.941Improved
Use ComplEx instead of RESCAL on a large relation setknowledge graph embedding · FB15Kfiltered MRR: 0.461 -> 0.692Improved
Use ComplEx instead of RESCAL on a fully observed data set (Kinships, UMLS)knowledge graph embedding · Kinships and UMLSaverage precisionNo clear effect
Increase negatives per positive from 1 to 10 (ablation)ComplEx · FB15Kfiltered MRR: up to +0.08Improved
Increase negatives per positive from 1 to 100 (ablation)ComplEx · FB15Kfiltered MRR: 0.737 filtered MRR and 64.8% Hits@1 at 100 negativesImproved
Increase negatives per positive from 100 to 200 (ablation)ComplEx · FB15Kfiltered MRR: decreases at 200 negativesHurt
Increase negatives per positive beyond one (ablation)ComplEx · WN18filtered MRRNo clear effect
Add L2 regularization (optimal lambda instead of 0) (ablation)ComplEx · WN18 and FB15Kfiltered MRR: up to +0.05Improved
Increase rank from 50 to 200 (ablation)ComplEx · WN18 and FB15Kfiltered MRR: little performance gain for ranks 50-200No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy