Graph Disentangled Contrastive Learning with Personalized Transfer for Cross-Domain Recommendation

Authors: Jing Liu, Lele Sun, Weizhi Nie, Peiguang Jing, Yuting Su

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on four real-world datasets demonstrate the superiority of GDCCDR over state-of-the-art methods.
Researcher Affiliation Academia Jing Liu, Lele Sun, Weizhi Nie, Peiguang Jing, Yuting Su* School of Electrical and Information Engineering, Tianjin University, China {jliu tju, sunlele, weizhinie, pgjing, ytsu}@tju.edu.cn
Pseudocode No The paper describes its methods using mathematical formulations and textual descriptions but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide an explicit statement or link indicating that the source code for the methodology is openly available.
Open Datasets Yes We evaluate GDCCDR on the Amazon dataset1, specifically Sport&Phone, Sport&Cloth, Elec&Phone, and Elec&Cloth. 1http://jmcauley.ucsd.edu/data/amazon/index 2014.html
Dataset Splits No The paper mentions training and test sets and a leave-one-out strategy but does not explicitly provide percentages or counts for distinct training, validation, and test splits.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions 'Py Torch implementation' but does not provide specific version numbers for software dependencies like PyTorch or Python.
Experiment Setup Yes The embedding dimension (d) is set to 128 for all methods, with a fixed learning rate of 0.001, a batch size of 1024, and a dropout rate of 0.5. The low-rank (k) is 10, the proximate temperature (τp) is 0.05, the L2 regularization coefficient (λl) is selected from {0.05, 0.005, 0.0005}. The final embeddings of GNN-based methods are obtained through mean pooling. For point-wise loss, we have four negative samples per positive sample.