{"url":"/method/gatv2","slug":"gatv2","name":"GATv2","full_name":"Graph Attention Network v2","full_name_withheld":false,"description_markdown":"The __GATv2__ operator from the [“How Attentive are Graph Attention Networks?”](https://arxiv.org/abs/2105.14491) paper, which fixes the static attention problem of the standard [GAT](https://paperswithcode.com/method/gat) layer: since the linear layers in the standard GAT are applied right after each other, the ranking of attended nodes is unconditioned on the query node. In contrast, in GATv2, every node can attend to any other node.\r\n\r\nGATv2 scoring function:\r\n\r\n$e_{i,j} =\\mathbf{a}^{\\top}\\mathrm{LeakyReLU}\\left(\\mathbf{W}[\\mathbf{h}_i \\, \\Vert \\,\\mathbf{h}_j]\\right)$","description_state":"present","introduced_year":null,"introduced_by":{"title":"How Attentive are Graph Attention Networks?","paper":"/paper/how-attentive-are-graph-attention-networks","first_author":"Shaked Brody","n_authors":3,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/how-attentive-are-graph-attention-networks"},"source":{"url":"https://arxiv.org/abs/2105.14491v3","title":"How Attentive are Graph Attention Networks?","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"Graphs","area_id":"graphs","collection":"Graph Models","url":"/methods/category/graph-models","pwc_aliases":[]}],"n_papers_tagged":9,"archive_num_papers":9,"papers_newest_first":[{"paper":"/paper/graph-neural-networks-in-supply-chain","title":"Graph Neural Networks in Supply Chain Analytics and Optimization: Concepts, Perspectives, Dataset and Benchmarks","date":"2024-11-13","arxiv_id":"2411.08550","n_code_links":3,"syntology":null},{"paper":"/paper/perceived-text-relevance-estimation-using","title":"Perceived Text Relevance Estimation Using Scanpaths and GNNs","date":"2024-11-04","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":null,"title":"GATher: Graph Attention Based Predictions of Gene-Disease Links","date":"2024-09-23","arxiv_id":"2409.16327","n_code_links":0,"syntology":null},{"paper":"/paper/on-the-power-of-graph-neural-networks-and","title":"On the Power of Graph Neural Networks and Feature Augmentation Strategies to Classify Social Networks","date":"2024-01-11","arxiv_id":"2401.06048","n_code_links":1,"syntology":null},{"paper":"/paper/optimization-and-interpretability-of-graph","title":"Optimization and Interpretability of Graph Attention Networks for Small Sparse Graph Structures in Automotive Applications","date":"2023-05-25","arxiv_id":"2305.16196","n_code_links":1,"syntology":null},{"paper":null,"title":"Gradient Derivation for Learnable Parameters in Graph Attention Networks","date":"2023-04-21","arxiv_id":"2304.10939","n_code_links":0,"syntology":null},{"paper":"/paper/a-new-graph-node-classification-benchmark","title":"A New Graph Node Classification Benchmark: Learning Structure from Histology Cell Graphs","date":"2022-11-11","arxiv_id":"2211.06292","n_code_links":1,"syntology":{"ran":0,"of":3,"unverified":3,"pointer_only":0}},{"paper":null,"title":"Distance-Geometric Graph Attention Network (DG-GAT) for 3D Molecular Geometry","date":"2022-07-16","arxiv_id":"2207.08023","n_code_links":0,"syntology":null},{"paper":"/paper/how-attentive-are-graph-attention-networks","title":"How Attentive are Graph Attention Networks?","date":"2021-05-30","arxiv_id":"2105.14491","n_code_links":8,"syntology":{"ran":6,"of":16,"unverified":10,"pointer_only":1}}],"papers_shown":9,"tasks":[{"task":"/task/graph-attention","name":"Graph Attention","papers":5},{"task":"/task/graph-classification","name":"Graph Classification","papers":2},{"task":"/task/graph-neural-network","name":"Graph Neural Network","papers":2},{"task":"/task/anomaly-detection","name":"Anomaly Detection","papers":1},{"task":"/task/deep-learning","name":"Deep Learning","papers":1},{"task":"/task/demand-forecasting","name":"Demand Forecasting","papers":1},{"task":"/task/drug-discovery","name":"Drug Discovery","papers":1},{"task":"/task/graph-learning","name":"Graph Learning","papers":1},{"task":"/task/graph-property-prediction","name":"Graph Property Prediction","papers":1},{"task":"/task/graph-regression","name":"Graph Regression","papers":1},{"task":"/task/link-prediction","name":"Link Prediction","papers":1},{"task":"/task/molecular-property-prediction","name":"Molecular Property Prediction","papers":1},{"task":"/task/node-classification","name":"Node Classification","papers":1},{"task":"/task/node-property-prediction","name":"Node Property Prediction","papers":1},{"task":"/task/product-categorization","name":"Product Categorization","papers":1},{"task":"/task/product-relation-classification","name":"Product Relation Classification","papers":1},{"task":"/task/product-relation-detection","name":"Product Relation Detection","papers":1},{"task":"/task/production-forecasting","name":"Production Forecasting","papers":1},{"task":"/task/representation-learning","name":"Representation Learning","papers":1},{"task":"/task/specificity","name":"Specificity","papers":1}],"tasks_shown":20,"n_tasks":21,"usage_by_year":[{"year":"2021","papers":1},{"year":"2022","papers":2},{"year":"2023","papers":2},{"year":"2024","papers":4}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/gatv2"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}