CPa-WAC: Constellation Partitioning-based Scalable Weighted Aggregation Composition for Knowledge Graph Embedding

Authors: Sudipta Modak, Aakarsh Malhotra, Sarthak Malik, Anil Surisetty, Esam Abdel-Raheem

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The results from our experiments on standard databases, such as Wordnet and Freebase, show that by achieving meaningful partitioning, any knowledge graph can be broken down into subgraphs and processed separately to learn embeddings.
Researcher Affiliation Collaboration Sudipta Modak1,2 , Aakarsh Malhotra2 , Sarthak Malik2 , Anil Surisetty2 , Esam Abdel-Raheem1 1Department of Electrical and Computer Engineering, University of Windsor, ON, Canada 2AI Garage, Mastercard, Gurugram, Haryana, India
Pseudocode Yes Algorithm 1 Louvain Constellation Partitioning
Open Source Code Yes The code is made available for the research community1, and summarized in Algorithm 1. 1https://github.com/ganzagun/CPa-WAC
Open Datasets Yes As summarized in Table 1, we use the most widely used Wordnet [Miller, 1995] and Freebase [Bollacker et al., 2008] for our experiments.
Dataset Splits Yes Table 1: Dataset E R Training Valid Test
Hardware Specification Yes All experiments have been conducted on an I7-13700, 2.1 GHz system with 32 GB RAM and NVIDIA RTX A2000 12 GB GPU.
Software Dependencies No The paper mentions using "Adam W optimizer" and libraries like "Pytorch-Biggraph" and "DGL-KE" but does not specify version numbers for any of the software dependencies used in their implementation.
Experiment Setup Yes The Adam W optimizer is utilized to train the weights of the proposed architecture for a total of 400 epochs for all partition-based experimentation. Furthermore, all state-of-the-art models have been implemented using the same hyperparameter settings as the proposed architecture. This setting includes batch number, regularization rate, weight decay, learning rate, and the same embedding dimensions for each model.