Rot-Pro: Modeling Transitivity by Projection in Knowledge Graph Embedding

Authors: Tengwei Song, Jie Luo, Lei Huang

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that the proposed Rot-Pro model effectively learns the transitivity pattern and achieves the state-of-the-art results on the link prediction task in the datasets containing transitive relations.
Researcher Affiliation Academia Tengwei Song, Jie Luo , Lei Huang State Key Laboratory of Software Development Environment Beihang University, Beijing, 100191 {songtengwei,luojie,huangleiai}@buaa.edu.cn
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statement or link indicating that the source code for the methodology is openly available.
Open Datasets Yes We evaluate the Rot-Pro model on four well-known benchmarks. In general, FB15k-237 and WN18RR are two widely-used benchmarks and YAGO3-10 and Countries are two benchmarks with abundant relation patterns including transitivity. ... FB15k-237 [24] ... WN18RR [23] ... YAGO3-10 [13] ... Countries [8]
Dataset Splits No The paper states that "corrupt triples that appear in training, validation, or test sets are removed during ranking." indicating the use of a validation set. However, it does not provide specific details on the validation split (e.g., percentages or sample counts), nor does it reference predefined splits with such information.
Hardware Specification No The paper does not provide any specific details regarding the hardware used for running the experiments (e.g., GPU models, CPU types, or memory).
Software Dependencies No The paper does not specify any software dependencies with version numbers.
Experiment Setup Yes With the hyper-parameters introduced, we train Rot-Pro using a grid search of hyper-parameters: fixed margin γ in Equation 9 {0.1, 4.0, 6.0, 9.0, 16.0, 20.0}, weights tuning hyper-parameters for loss, α {0.0001, 0.0005, 0.0008}, value of γm in Equation 10 {1e 6, 5e 6, 1e 5}, value of β in Equation 10 {1.3, 1.5, 2.0}.