Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity

Bai, Yunsheng; Ding, Hao; Qiao, Yang; Marinovic, Agustin; Gu, Ken; Chen, Ting; Sun, Yizhou; Wang, Wei

doi:10.48550/arxiv.1904.01098

Cited by 13 publications

(14 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where is a hyper-parameter controlling the contributions of the two parts. Note that, our solution is different from existing graph neural networks [3,26] in two aspects. First, our group representation learning network is hierarchical, which first learns the group members' personal preferences in the lower-layer (i.e., IPM) and then infers groups' representations in the higher-layer (i.e., HRL).…”

Section: Hyperedge Embedding-based Group Representation Learningmentioning

confidence: 83%

“…To integrate the group similarity based on common members in the learning process of hyperedge embedding, we devise a GNN-based hyperedge embedding model, called HRL. As OGs tend to be formed by chance [25,65], HRL also adopts an inductive graph embedding method [3,26] as its building block to generate embeddings for hyperedges, where a weighted feature aggregation scheme is proposed to account for the similarity between two groups.…”

Section: Hyperedge Embedding-based Group Representation Learningmentioning

confidence: 99%

See 1 more Smart Citation

Hierarchical Hyperedge Embedding-based Representation Learning for Group Recommendation

Guo

Yin

Chen

et al. 2021

Preprint

View full text Add to dashboard Cite

Group Recommendation (GR) aims to recommend items to a group of users. In this work, we study GR in a particular scenario, namely Occasional Group Recommendation (OGR), where groups are formed ad-hoc and users may just constitute a group for the first time, that is, the historical group-item interaction records are highly limited. Most state-of-the-art works have addressed the challenge by aggregating group members' personal preferences to learn the group representation. However, the representation learning for a group is most complex beyond the aggregation or fusion of group member representation, as the personal preferences and group preferences may be in different spaces and even orthogonal. In addition, the learned user representation is not accurate due to the sparsity of users' interaction data. Moreover, the group similarity in terms of common group members has been overlooked, which however has the great potential to improve the group representation learning. In this work, we focus on addressing the aforementioned challenges in group representation learning task, and devise a hierarchical hyperedge embedding-based group recommender, namely HyperGroup. Specifically, we propose to leverage the user-user interactions to alleviate the sparsity issue of user-item interactions, and design a Graph Neural Network (GNN)-based representation learning network to enhance the learning of individuals' preferences from their friends' preferences, which provides a solid foundation for learning groups' preferences. To exploit the group similarity (i.e., overlapping relationships among groups) to learn a more accurate group representation from highly limited group-item interactions, we connect all groups as a network of overlapping sets (a.k.a. hypergraph), and treat the task of group preference learning as embedding hyperedges (i.e., user sets/groups) in a hypergraph, where an inductive hyperedge embedding method is proposed. To further enhance the group-level preference modeling, we develop a joint training strategy to learn both user-item and group-item interactions in the same process. We conduct extensive experiments on two real-world datasets and the experimental results demonstrate the superiority of our proposed HyperGroup in comparison to the state-of-the-art baselines.

show abstract

Section: Hyperedge Embedding-based Group Representation Learningmentioning

confidence: 83%

Section: Hyperedge Embedding-based Group Representation Learningmentioning

confidence: 99%

Hierarchical Hyperedge Embedding-based Representation Learning for Group Recommendation

Guo

Yin

Chen

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Figure 4 contrasts the original point cloud with the points reconstructed from each pooling method while the connectivity is unchanged; the figures for all point clouds are available in the supplementary Table 3: MSE (values in scale of 10 −3 ) in the autoencoder experiment. 4 The Rank row indicates the average ranking of the methods across all datasets. material.…”

Section: Discussionmentioning

confidence: 99%

“…Other notable works include those of Diehl [17], Bodnar et al [8], and a broad group of global pooling methods (cf. Section 4) that reduce a whole graph to a single vector [33,58,51,41,13,2,54,4].…”

Section: Pooling In Graph Neural Networkmentioning

confidence: 99%

Understanding Pooling in Graph Neural Networks

Grattarola¹,

Zambon²,

Bianchi³

et al. 2021

Preprint

View full text Add to dashboard Cite

Inspired by the conventional pooling layers in convolutional neural networks, many recent works in the field of graph machine learning have introduced pooling operators to reduce the size of graphs. The great variety in the literature stems from the many possible strategies for coarsening a graph, which may depend on different assumptions on the graph structure or the specific downstream task. In this paper we propose a formal characterization of graph pooling based on three main operations, called selection, reduction, and connection, with the goal of unifying the literature under a common framework. Following this formalization, we introduce a taxonomy of pooling operators and categorize more than thirty pooling methods proposed in recent literature. We propose criteria to evaluate the performance of a pooling operator and use them to investigate and contrast the behavior of different classes of the taxonomy on a variety of tasks.Preprint. Under review.

show abstract

“…After feature graphs of both modalities are extracted, we embed them into vector forms using an attention mechanism called multi-scale node attention [31]. Given a graph G having N nodes {h 1 , ..., h N } and M edges {r 1 , ..., r M } where h i and r j ∈ R d is a feature of a node and an edge accordingly, the embedded vector of G, noted as a G ∈ R 2d , is obtained by the following formula:…”

Section: Graph Embeddingmentioning

confidence: 99%

A Deep Local and Global Scene-Graph Matching for Image-Text Retrieval

Nguyen¹,

Nguyen²,

Gurrin³

2021

Preprint

View full text Add to dashboard Cite

Conventional approaches to image-text retrieval mainly focus on indexing visual objects appearing in pictures but ignore the interactions between these objects. Such objects occurrences and interactions are equivalently useful and important in this field as they are usually mentioned in the text. Scene graph presentation is a suitable method for the image-text matching challenge and obtained good results due to its ability to capture the inter-relationship information. Both images and text are represented in scene graph levels and formulate the retrieval challenge as a scene graph matching challenge. In this paper, we introduce the Local and Global Scene Graph Matching (LGSGM) model that enhances the state-of-the-art method by integrating an extra graph convolution network to capture the general information of a graph. Specifically, for a pair of scene graphs of an image and its caption, two separate models are used to learn the features of each graph's nodes and edges. Then a Siamese-structure graph convolution model is employed to embed graphs into vector forms. We finally combine the graph-level and the vectorlevel to calculate the similarity of this image-text pair. The empirical experiments show that our enhancement with the combination of levels can improve the performance of the baseline method by increasing the recall by more than 10% on the Flickr30k dataset.

show abstract

Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity

Cited by 13 publications

References 33 publications

Hierarchical Hyperedge Embedding-based Representation Learning for Group Recommendation

Hierarchical Hyperedge Embedding-based Representation Learning for Group Recommendation

Understanding Pooling in Graph Neural Networks

A Deep Local and Global Scene-Graph Matching for Image-Text Retrieval

Contact Info

Product

Resources

About