10,890 Hits in 3.1 sec

Relphormer: Relational Graph Transformer for Knowledge Graph Representation [article]

Zhen Bi, Siyuan Cheng, Ningyu Zhang, Xiaozhuan Liang, Feiyu Xiong, Huajun Chen
2022 arXiv   pre-print
To this end, we propose a new variant of Transformer for knowledge graph representation dubbed Relphormer.  ...  Moreover, we propose masked knowledge modeling as a new paradigm for knowledge graph representation learning to unify different link prediction tasks.  ...  Table 5 : 5 Inference efficiency comparison. |E |, |R| and |T | are numbers of entities, relations, inference samples in the graph respectively. 𝑘 is the length of input sequence.  ... 
arXiv:2205.10852v2 fatcat:7iw263mzxzehrhx4ogjadzc5mi

FLOWGEN: Fast and slow graph generation [article]

Aman Madaan, Yiming Yang
2022 arXiv   pre-print
We present FLOWGEN, a graph-generation model inspired by the dual-process theory of mind that generates large graphs incrementally.  ...  Experiments on real-world graphs show that ours can successfully generate graphs similar to those generated by a single large model in a fraction of time.  ...  FLOWGEN leverages this mismatch in problem difficulty by dynamically switching from a small (FAST) model to a large (SLOW) model for efficient graph generation.  ... 
arXiv:2207.07656v2 fatcat:lnhcavvrzbcwlnaigmozk33xby

GraphHINGE: Learning Interaction Models of Structured Neighborhood on Heterogeneous Information Network [article]

Jiarui Jin, Kounianhua Du, Weinan Zhang, Jiarui Qin, Yuchen Fang, Yong Yu, Zheng Zhang, Alexander J. Smola
2021 arXiv   pre-print
Recent attempts either rely on explicit path reachability to leverage path-based semantic relatedness or graph neighborhood to learn heterogeneous network representations before predictions.  ...  Next, in order to address the complexity issue on large-scale networks, we formulate the interaction modules via a convolutional framework and learn the parameters efficiently with fast Fourier transform  ...  'An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph'.  ... 
arXiv:2011.12683v2 fatcat:nv27cwdpkrblto3fja7wqdnhdy

Text Enriched Sparse Hyperbolic Graph Convolutional Networks [article]

Nurendra Choudhary, Nikhil Rao, Karthik Subbian, Chandan K. Reddy
2022 arXiv   pre-print
These extracted features in conjunction with semantic features from the language model (for robustness) are used for the final downstream task.  ...  Graph Neural Networks (GNNs) and their hyperbolic variants provide a promising approach to encode such networks in a low-dimensional latent space through neighborhood aggregation and hierarchical feature  ...  to leverage the adjacency tensor's sparsity in the hyperbolic space for computational efficiency.  ... 
arXiv:2207.02368v2 fatcat:blf425ppw5c6zbndc7uiyhzkey

Adaptive Neighborhood Graph Construction for Inference in Multi-Relational Networks [article]

Shobeir Fakhraei, Dhanya Sridhar, Jay Pujara, Lise Getoor
2016 arXiv   pre-print
This process often consists of generating a candidate graph, pruning the candidate graph to make a neighborhood graph, and then performing inference on the variables (i.e., nodes) in the neighborhood graph  ...  A neighborhood graph, which represents the instances as vertices and their relations as weighted edges, is the basis of many semi-supervised and relational models for node labeling and link prediction.  ...  Our framework enables rich models for active inference that interleave inference and neighborhood graph construction, efficiently using multi-relational data while maintaining scalable performance.  ... 
arXiv:1607.00474v1 fatcat:j6qez4542va7fcyqw3qf34qhsu

Collaboration-Aware Graph Convolutional Networks for Recommendation Systems [article]

Yu Wang, Yuying Zhao, Yi Zhang, Tyler Derr
2022 arXiv   pre-print
First, we theoretically analyze how message-passing captures and leverages the collaborative effect in predicting user preferences.  ...  Therefore, in this work we aim to demystify the collaborative effect captured by message-passing in GNNs and develop new insights towards customizing message-passing for recommendation.  ...  On one hand, graph-based models could leverage neighborhood information to augment the weak supervision for low-degree (cold-start) nodes.  ... 
arXiv:2207.06221v1 fatcat:52kmotqmefgqncmeuq3y6dj224

Scaling Graph Propagation Kernels for Predictive Learning

Priyesh Vijayan, Yash Chandak, Mitesh M. Khapra, Srinivasan Parthasarathy, Balaraman Ravindran
2022 Frontiers in Big Data  
I-HOP scales differentiable graph kernels to capture and summarize information from a larger neighborhood in each iteration by leveraging a historical neighborhood summary obtained in the previous iteration  ...  To perform downstream analysis on such data, it is crucial to capture relational information of nodes over their expanded neighborhood efficiently.  ...  Thus, I-HOP is more generic, flexible, CPU RAM efficient over GNNAutoscale and can leverage useful label information.  ... 
doi:10.3389/fdata.2022.616617 pmid:35464122 pmcid:PMC9026185 fatcat:7qhsnr3elzcgnoq6sfw4hixrui

An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph [article]

Jiarui Jin, Jiarui Qin, Yuchen Fang, Kounianhua Du, Weinan Zhang, Yong Yu, Zheng Zhang, Alexander J. Smola
2020 arXiv   pre-print
In this paper, we propose an end-to-end Neighborhood-based Interaction Model for Recommendation (NIRec) to address the above problems.  ...  To the best of our knowledge, this is the first work providing an efficient neighborhood-based interaction model in the HIN-based recommendations.  ...  LINE [28] proposed an edge-sampling algorithm improving both the effectiveness and the efficiency of the inference.  ... 
arXiv:2007.00216v1 fatcat:cruhwo3hdzdmpe5v3cvdboafaa

Inductive Graph Neural Networks For Transfer Learning

Atlas Wang
2022 Zenodo  
"head" nodes in the same graph yet with much richer neighborhood information.  ...  in the second graph (target), our aim is to predict links in the target graph by leveraging the source graph's richer information.  ...  ., product type) and behavioral signals Cold or Isolated Learning efficient mapping from features to fill the structural gap Error from feature-structure correlation Large error for cold start nodes  ... 
doi:10.5281/zenodo.6501657 fatcat:heekjiz4djbrhhcc75xsdxhiie

Graphical Games for UAV Swarm Control Under Time-Varying Communication Networks [article]

Malintha Fernando, Ransalu Senanayake, Ariful Azad, Martin Swany
2022 arXiv   pre-print
Finally, we discuss extending the proposed framework toward general-sum stochastic games by leveraging deep Q-learning and model-predictive control.  ...  We present a general-sum, factorizable payoff function for cooperative UAV swarms based on the aggregated local states and yield a Nash equilibrium for the stage games.  ...  Our solution stems from the idea that, we can solve each Γ i in an independent and a decentralized manner where each robot infers the best response actions for its neighborhood by leveraging the local  ... 
arXiv:2205.02203v1 fatcat:tuvrt7klbnfmdbgc3mqwpjy2lm

LinkNBed: Multi-Graph Representation Learning with Entity Linkage

Rakshit Trivedi, Bunyamin Sisman, Xin Luna Dong, Christos Faloutsos, Jun Ma, Hongyuan Zha
2018 Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
An effective and scalable approach to jointly learn over multiple graphs and eventually construct a unified graph is a crucial next step for the success of knowledge-based inference for many downstream  ...  We identify entity linkage across graphs as a vital component to achieve our goal. We design a novel objective that leverage entity linkage and build an efficient multi-task training procedure.  ...  We thank the reviewers for their valuable comments and efforts towards improving our manuscript. This project was supported in part by NSF(IIS-1639792, IIS-1717916).  ... 
doi:10.18653/v1/p18-1024 dblp:conf/acl/FaloutsosTSDMZ18 fatcat:2jbo23d3d5benelgik2byae3ni

Graph Convolutional Neural Networks for Web-Scale Recommender Systems

Rex Ying, Ruining He, Kaifeng Chen, Pong Eksombatchai, William L. Hamilton, Jure Leskovec
2018 Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD '18  
We also develop an efficient MapReduce model inference algorithm to generate embeddings using a trained model.  ...  We develop a data-efficient Graph Convolutional Network (GCN) algorithm PinSage, which combines efficient random walks and graph convolutions to generate embeddings of nodes (i.e., items) that incorporate  ...  Kulikov for help in nearest neighbor query of the item embeddings.  ... 
doi:10.1145/3219819.3219890 dblp:conf/kdd/YingHCEHL18 fatcat:xp5aezpyjbcvfdjjmke3ivjgm4

Graph Neural Networks Intersect Probabilistic Graphical Models: A Survey [article]

Chenqing Hua
2022 arXiv   pre-print
Then we discuss how GNNs are implemented in PGMs for more efficient inference and structure learning.  ...  Graph Neural Networks (GNNs) are new inference methods developed in recent years and are attracting growing attention due to their effectiveness and flexibility in solving inference and learning problems  ...  Conversely, GNNs can be used in PGMs for two purposes: (1) GNNs can efficiently and effectively leverage inference in PGMs; (2) GNNs can enhance the directed acyclic graph learning in PGMs.  ... 
arXiv:2206.06089v1 fatcat:twv6sh5f3ngkrjfiyoger6vuuq

HOPF: Higher Order Propagation Framework for Deep Collective Classification [article]

Priyesh Vijayan, Yash Chandak, Mitesh M. Khapra, Srinivasan Parthasarathy, Balaraman Ravindran
2018 arXiv   pre-print
Recent state-of-the-art models for CC learn end-to-end differentiable variations of Weisfeiler-Lehman (WL) kernels to aggregate multi-hop neighborhood information.  ...  In this work, we propose a Higher Order Propagation Framework, HOPF, which provides an iterative inference mechanism for these powerful differentiable kernels.  ...  To make this process more efficient, propagation kernels (Neumann et al. 2016) provide additional schemes for diffusing the available information across the graph.  ... 
arXiv:1805.12421v6 fatcat:hwisfq5cdnddpcota2xdh5ni4u

Learning to Cluster Faces via Confidence and Connectivity Estimation [article]

Lei Yang, Dapeng Chen, Xiaohang Zhan, Rui Zhao, Chen Change Loy, Dahua Lin
2020 arXiv   pre-print
Face clustering is an essential tool for exploiting the unlabeled face data, and has a wide range of applications including face annotation and retrieval.  ...  However, they usually involve heuristic steps and require numerous overlapped subgraphs, severely restricting their accuracy and efficiency.  ...  Empirically, a 1-layer GCN takes 37G CPU Ram and 92s with 16 CPU on a graph with 5.2M vertices for inference.  ... 
arXiv:2004.00445v2 fatcat:r7g6wywjjrhwbnl7hhj3xos2zi
« Previous Showing results 1 — 15 out of 10,890 results