WebApr 1, 2024 · This paper proposes a new social recommendation system based on a light graph convolution network, called ’SocialLGN’. SocialLGN innovatively extends the user/item representation propagation mechanism in LightGCN to incorporate two graphs (i.e., the user-item interaction graph and social graph). WebJul 25, 2024 · LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation Pages 639–648 ABSTRACT References Cited By Index Terms …
Implement Your Own Music Recommender with Graph Neural Networks (LightGCN)
WebFeb 6, 2024 · LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation Xiangnan He, Kuan Deng, +3 authors Meng Wang Published 6 February 2024 Computer Science Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval WebApr 14, 2024 · In this paper, we first analyze the limitations of LightGCN, which is a representative work of GCN-based CF. Based on empirical studies, we reveal that LightGCN is time-consuming in training and suffers from the scale oscillation issue. Then, we propose a novel Accelerated Light Graph Convolution Network (ALGCN) for collaborative filtering. ccthumstudy
SocialLGN: Light graph convolution network for social recommendation
WebJul 25, 2024 · LightGCN is an improvement over NGCF [29] which was shown to outperform many previous models such as graph-based GC-MC [35] and PinSage [34], neural … WebOct 28, 2024 · LightGCN makes an early attempt to simplify GCNs for collaborative filtering by omitting feature transformations and nonlinear activations. In this paper, we take one step further to propose an ultra-simplified formulation of GCNs (dubbed UltraGCN), which skips infinite layers of message passing for efficient recommendation. WebDec 13, 2024 · LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation, Paper in arXiv. Author: Prof. Xiangnan He (staff.ustc.edu.cn/~hexn/) ... ├── analytics // code for all the analytics ops and utils ├── code // code dir for LightGCN ├── data // pre-processed data for the training ops ├── dataloader ... butchers ashby de la zouch