Rule-Guided Compositional Representation Learning on Knowledge Graphs
Authors: Guanglin Niu, Yongfei Zhang, Bo Li, Peng Cui, Si Liu, Jingyang Li, Xiaowei Zhang2950-2958
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results illustrate that RPJE outperforms other state-of-the-art baselines on KG completion task, which also demonstrate the superiority of utilizing logic rules as well as paths for improving the accuracy and explainability of representation learning. |
| Researcher Affiliation | Academia | 1Beijing Key Laboratory of Digital Media, School of Computer Science and Engineering, Beihang University, 2State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, 3Department of Computer Science and Technology, Tsinghua University, 4College of Computer Science and Technology, Qingdao University |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. It mentions implementing baselines from their source codes, but not their own proposed method. |
| Open Datasets | Yes | We evaluate our model on four typical datasets: FB15K and FB15K-237 both extracted from the large-scale Freebase (Bollacker, Gottlob, and Flesca 2008), WN18 extracted from Word Net (Miller 1995) and NELL-995 extracted from NELL (Mitchell et al. 2018). |
| Dataset Splits | Yes | Table 3: Statistics of datasets used in the experiments. Dataset #Rel #Ent #Train #Valid #Test FB15K 1,345 14,951 483,142 50,000 59,071 FB15K-237 237 14,541 272,115 17,535 20,466 WN18 18 40,943 141,442 5,000 5,000 NELL-995 200 75,492 123,370 15,000 15,838 |
| Hardware Specification | No | The paper does not provide specific hardware details (like GPU or CPU models, or memory specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions using tools like AMIE+ and optimization methods like SGD, but does not provide specific version numbers for any ancillary software, libraries, or frameworks used in the experiments. |
| Experiment Setup | Yes | To guarantee fair comparison, we adopt the following evaluation settings in our work: (1) 100 mini-batches are created on datasets. (2) The entity and relation embeddings are initialized randomly and limited to unit vectors. (3) Following the same configurations as many prevailing baselines, the learning rate is chosen as 0.001, γ1 and γ2 are selected as γ1 = γ2 = 1, the embedding dimension is set to 50 for WN18 and 100 for other three datasets considering only 18 relations exist in WN18, dissimilarity is selected as L1 and training epochs is set to 500. In addition, we employ a grid search to select the other optimal hyper-parameters. We manually tune the margin γ3 in {1, 1, 5, 2, 2.5, 3}, and the weight coefficients α1, α2 both in {0, 5, 1, 1.5, 2, 3, 5}. The best models are selected on validation sets. The resulting optimal of margin γ3 and the weight coefficients α1, α2 are assigned to: γ3 = 1, α1 = 1, α2 = 3. |