LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation

Authors: Xuheng Cai, Chao Huang, Lianghao Xia, Xubin Ren

ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments conducted on several benchmark datasets demonstrate the significant improvement in performance of our model over the state-of-the-arts. Further analyses demonstrate the superiority of Light GCL s robustness against data sparsity and popularity bias. The source code of our model is available at https://github.com/HKUDS/Light GCL. To verify the superiority and effectiveness of the proposed Light GCL method, we perform extensive experiments to answer the following research questions:
Researcher Affiliation Academia Xuheng Cai Chao Huang Lianghao Xia Xubin Ren Department of Computer Science, University of Hong Kong {rickcai, lhaoxia}@hku.hk chaohuang75gmail.com xubinrencs@gmail.com
Pseudocode No The paper describes the model with mathematical equations (e.g., Eq. 1-8) and textual explanations, but it does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code Yes The source code of our model is available at https://github.com/HKUDS/Light GCL.
Open Datasets Yes We evaluate our model and the baselines on five real-world datasets: Yelp (29,601 users, 24,734 items, 1,517,326 interactions): a dataset collected from the rating interactions on Yelp platform; Gowalla (50,821 users, 57,440 items, 1,172,425 interactions): a dataset containing users check-in records collected from Gowalla platform; ML-10M (69,878 users, 10,195 items, 9,988,816 interactions): a well-known movie-rating dataset for collaborative filtering; Amazon-book (78,578 users, 77,801 items, 2,240,156 interactions): a dataset composed of users ratings on books collected from Amazon; and Tmall (47,939 users, 41,390 items, 2,357,450 interactions): a E-commerce dataset containing users purchase records on different products in Tmall platform.
Dataset Splits Yes In accordance with He et al. (2020) and Wu et al. (2021), we split the datasets into training, validation and testing sets with a ratio of 7:2:1.
Hardware Specification No The paper does not provide specific details regarding the hardware used for running the experiments, such as GPU models, CPU types, or memory specifications.
Software Dependencies No The paper does not explicitly list specific software dependencies with version numbers (e.g., Python version, library versions like PyTorch or TensorFlow) required to reproduce the experiments.
Experiment Setup Yes the embedding size is set as 32; the batch size is 256; two convolutional layers are used for GCN models. For our Light GCL, the regularization weights λ1 and λ2 are tuned from {1e-5, 1e-6, 1e-7} and {1e4, 1e-5}, respectively. The temperature τ is searched from {0.3, 0.5, 1, 3 ,10}. The dropout rate is chosen from {0, 0.25}. The rank (i.e., q) for SVD, is set as 5. We use the Adam optimizer with a learning rate of 0.001 decaying at the rate of 0.98 until the rate reaches 0.0005.