Are Graph Augmentations Necessary?

Yu, Junliang; Yin, Hongzhi; Xia, Xin; Chen, Tong; Cui, Lizhen; Nguyen, Quoc Viet Hung

doi:10.1145/3477495.3531937

Cited by 208 publications

(132 citation statements)

References 16 publications

Supporting

Mentioning

132

Contrasting

Order By: Relevance

“…SGL [34] models user-item interactions as a graph and builds multiple views by conducting node dropout, edge dropout, and random walk on the user-item graph. Similarly, Similarly, Yu et al [40] propose a simple CL method to perturb the graph embedding space with uniform noise to build contrastive views for the training of recommendation model. At the same time, some researchers model user-item interactions with hypergraph, and conduct CL over augmented hypergraph views for the user/item representation learning [35,39,42].…”

Section: Contrastive Learning In Recommendationmentioning

confidence: 99%

Contrastive Collaborative Filtering for Cold-Start Item Recommendation

Zhou,

Zhang,

Yang

2023

Preprint

View full text Add to dashboard Cite

The cold-start problem is a long-standing challenge in recommender systems. As a promising solution, content-based generative models usually project a cold-start item's content onto a warm-start item embedding to capture collaborative signals from item content so that collaborative filtering can be applied. However, since the training of the cold-start recommendation models is conducted on warm datasets, the existent methods face the issue that the collaborative embeddings of items will be blurred, which significantly degenerates the performance of cold-start item recommendation. To address this issue, we propose a novel model called Contrastive Collaborative Filtering for Cold-start item Recommendation (CCFCRec), which capitalizes on the co-occurrence collaborative signals in warm training data to alleviate the issue of blurry collaborative embeddings for cold-start item recommendation. In particular, we devise a contrastive collaborative filtering (CF) framework, consisting of a content CF module and a co-occurrence CF module to generate the content-based collaborative embedding and the co-occurrence collaborative embedding for a training item, respectively. During the joint training of the two CF modules, we apply a contrastive learning between the two collaborative embeddings, by which the knowledge about the co-occurrence signals can be indirectly transferred to the content CF module, so that the blurry collaborative embeddings can be rectified implicitly by the memorized co-occurrence collaborative signals during the applying phase. Together with the sound theoretical analysis, the extensive experiments conducted on real datasets demonstrate the superiority of the proposed model. The codes and datasets are available on https://github.com/zzhin/CCFCRec.

show abstract

Section: Contrastive Learning In Recommendationmentioning

confidence: 99%

Contrastive Collaborative Filtering for Cold-Start Item Recommendation

Zhou,

Zhang,

Yang

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The multimodal features are fed into different modality encoders. The modality encoders extract the representations and are general architectures used in other fields, such as ViT [13] for images and General [34] Coarse-grained Attention CL [40] Coarse-grained Attention None [6], [21] Fine-gained Attention None [30], [27], [57] Combined Attention None [44], [39] User-item Graph + Fine-gained Attention None [56] User-item Graph CL [59] Item-item Graph CL [58], [38] Item-item Graph None [33] Item-item Graph + Fine-gained Attention None [50], [45] Knowledge Graph None [2], [46] Knowledge Graph CL [8] Knowledge Graph + Fine-gained Attention None [43] Knowledge Graph + Filtration (graph) None [63], [55], [31] Filtration (graph) None [49], [4] MLP / Concat DRL [15], [28] Fine-gained Attention DRL [61], [36], [48] None DRL…”

Section: Procedures Of Mrsmentioning

confidence: 99%

Multimodal Recommender Systems: A Survey

Liu¹,

Hu²,

Xiao³

et al. 2023

Preprint

View full text Add to dashboard Cite

The recommender system (RS) has been an integral toolkit of online services. They are equipped with various deep learning techniques to model user preference based on identifier and attribute information. With the emergence of multimedia services, such as short video, news and etc., understanding these contents while recommending becomes critical. Besides, multimodal features are also helpful in alleviating the problem of data sparsity in RS. Thus, Multimodal Recommender System (MRS) has attracted much attention from both academia and industry recently. In this paper, we will give a comprehensive survey of the MRS models, mainly from technical views.First, we conclude the general procedures and major challenges for MRS. Then, we introduce the existing MRS models according to three categories, i.e., Feature Interaction, Feature Enhancement and Model Optimization. To make it convenient for those who want to research this field, we also summarize the dataset and code resources. Finally, we discuss some promising future directions of MRS and conclude this paper.

show abstract

“…The widely-used similarity functions include inner product [14] and neural network [11]. As suggested by recent work [19,33,36], the inner product supports highly efficient retrieval and usually exhibits stronger performance. Thus, for convenience, this work simply takes the representative inner product for analyses, i.e., model prediction can be expressed as ŷ𝑢𝑖 = e ⊤ 𝑢 e 𝑖 .…”

Section: Preliminariesmentioning

confidence: 99%

“…Meanwhile, we compare the model with various types of SOTA models, range from SGL [33], SimpleX [21], SimSGL [36], NCL [19]. We also report the distribution of the learned personalized 𝜏 for Adap-𝜏 and Cu-𝜏 as marked by the orange and blue regions.…”

Section: 12mentioning

confidence: 99%

See 1 more Smart Citation

Adap-$τ$: Adaptively Modulating Embedding Magnitude for Recommendation

Chen,

Wu,

et al. 2023

Preprint

View full text Add to dashboard Cite

Recent years have witnessed the great successes of embeddingbased methods in recommender systems. Despite their decent performance, we argue one potential limitation of these methods -the embedding magnitude has not been explicitly modulated, which may aggravate popularity bias and training instability, hindering the model from making a good recommendation. It motivates us to leverage the embedding normalization in recommendation. By normalizing user/item embeddings to a specific value, we empirically observe impressive performance gains (9% on average) on four real-world datasets. Although encouraging, we also reveal a serious limitation when applying normalization in recommendation -the performance is highly sensitive to the choice of the temperature 𝜏 which controls the scale of the normalized embeddings.To fully foster the merits of the normalization while circumvent its limitation, this work studied on how to adaptively set the proper 𝜏. Towards this end, we first make a comprehensive analyses of 𝜏 to fully understand its role on recommendation. We then accordingly develop an adaptive fine-grained strategy Adap-𝜏 for the temperature with satisfying four desirable properties including adaptivity, personalized, efficiency and model-agnostic. Extensive experiments have been conducted to validate the effectiveness of the proposal.

show abstract

Are Graph Augmentations Necessary?

Cited by 208 publications

References 16 publications

Contrastive Collaborative Filtering for Cold-Start Item Recommendation

Contrastive Collaborative Filtering for Cold-Start Item Recommendation

Multimodal Recommender Systems: A Survey

Adap-$τ$: Adaptively Modulating Embedding Magnitude for Recommendation

Contact Info

Product

Resources

About