A5057421310- Advanced Graph Neural Networks
- Recommender Systems and Techniques
- Topic Modeling
- Machine Learning in Materials Science
- Stochastic Gradient Optimization Techniques
- Image Retrieval and Classification Techniques
- Mental Health via Writing
University of Hong Kong
2023-2024
Stanford University
2023
Graph neural network (GNN) is a powerful learning approach for graph-based recommender systems. Recently, GNNs integrated with contrastive have shown superior performance in recommendation their data augmentation schemes, aiming at dealing highly sparse data. Despite success, most existing graph methods either perform stochastic (e.g., node/edge perturbation) on the user-item interaction graph, or rely heuristic-based techniques user clustering) generating views. We argue that these cannot...
Self-supervised learning (SSL) has gained significant interest in recent years as a solution to address the challenges posed by sparse and noisy data recommender systems. Despite growing number of SSL algorithms designed provide state-of-the-art performance various recommendation scenarios (e.g., graph collaborative filtering, sequential recommendation, social KG-enhanced recommendation), there is still lack unified frameworks that integrate across different domains. Such framework could...
Graph Neural Networks (GNNs) have demonstrated superior performance in various graph learning tasks, including recommendation, where they explore user-item collaborative filtering signals within graphs. However, despite their empirical effectiveness state-of-the-art recommender models, theoretical formulations of capability are scarce. Recently, researchers explored the expressiveness GNNs, demonstrating that message passing GNNs at most as powerful Weisfeiler-Lehman test, and combined with...
In recommendation systems, the cold-start issue is a long-standing problem where no historical interaction records are given for certain users or items. Under this circumstance, recommendations new items become challenging. To address problem, most existing approaches seek to discover latent common space and However, these methods require strong assumption that shared exists distributions of identical, which may limit performance. article, we propose novel model called Feature Matching...
Graph augmentation with contrastive learning has gained significant attention in the field of recommendation systems due to its ability learn expressive user representations, even when labeled data is limited. However, directly applying existing GCL models real-world environments poses challenges. There are two primary issues address. Firstly, lack consideration for noise can result noisy self-supervised signals, leading degraded performance. Secondly, many approaches rely on graph neural...
Self-supervised learning (SSL) has gained significant interest in recent years as a solution to address the challenges posed by sparse and noisy data recommender systems. Despite growing number of SSL algorithms designed provide state-of-the-art performance various recommendation scenarios (e.g., graph collaborative filtering, sequential recommendation, social KG-enhanced recommendation), there is still lack unified frameworks that integrate across different domains. Such framework could...
Graph Neural Networks (GNNs) have demonstrated superior performance on various graph learning tasks, including recommendation, where they leverage user-item collaborative filtering signals in graphs. However, theoretical formulations of their capability are scarce, despite empirical effectiveness state-of-the-art recommender models. Recently, research has explored the expressiveness GNNs general, demonstrating that message passing at most as powerful Weisfeiler-Lehman test, and combined with...