Papers of the day   All papers

Open Graph Benchmark: Datasets for Machine Learning on Graphs

Comments

Weihua Hu: Super excited to share Open Graph Benchmark (OGB)! OGB provides large-scale, diverse graph datasets to catalyze graph ML research. The datasets are easily accessible via OGB Python package with unified evaluation protocols and public leaderboards. Paper: https://arxiv.org/abs/2005.00687 https://t.co/5aGXqQxDJQ

2 replies, 432 likes


Weihua Hu: Excited to announce that "Open Graph Benchmark: Datasets for ML on Graphs" has been accepted to #NeurIPS as a spotlight! If you are working on Graph ML, please check it out for your next conference submission :) Website: https://ogb.stanford.edu Paper: https://arxiv.org/abs/2005.00687 https://t.co/khoC7MsidR

6 replies, 297 likes


Weihua Hu: Check out our 2nd major release of OGB! We have provided 5 new datasets, including a web-scale gigantic graph (with 100M+ nodes) and heterogeneous graphs. https://ogb.stanford.edu/

1 replies, 58 likes


Denny Britz: Stanford’s Open Graph Benchmark (OGB) is a collection of benchmark datasets, data loaders, and evaluators for graph machine learning: https://ogb.stanford.edu/ They even have a leaderboard :) Paper: https://arxiv.org/abs/2005.00687 https://t.co/96vehWu7Ux

0 replies, 46 likes


Marinka Zitnik: Subgraph Neural Networks https://arxiv.org/abs/2006.10538 GNNGuard: Defending GNNs against Adversarial Attacks https://arxiv.org/abs/2006.08149 Graph Meta Learning via Local Subgraphs https://arxiv.org/abs/2006.07889 Open Graph Benchmark: Datasets for ML on Graphs https://arxiv.org/abs/2005.00687

1 replies, 45 likes


Marinka Zitnik: We are thrilled to release the Open Graph Benchmark! OGB contains numerous biomedical datasets, including protein interaction nets, cross-species graphs, drug-drug interaction nets, and biomedical knowledge graphs #ML #graphs #networks

0 replies, 40 likes


Kristian Kersting: 🙏 for gathering this data. Glad that our previous @sfb876 work has helped to push this. Maybe, at some point, we can actually merge all graph datasets at a single location. Anyhow, great initiative. Thanks!

0 replies, 22 likes


Michele Catasta: Among the new tasks in this release, we included code summarization (AKA method naming): https://ogb.stanford.edu/docs/graphprop/#ogbg-code The dataset is based on the great work done by @github @miltos1 @mmjb86 for the CodeSearchNet challenge. Looking forward to your submissions!

0 replies, 11 likes


Weihua Hu: Our arXiv paper has also been updated: https://arxiv.org/abs/2005.00687 Includes detailed descriptions of each dataset, the extensive benchmark experiments, and discussion on research challenges and opportunities.

0 replies, 9 likes


Chaitanya Joshi: My favorite feature of the benchmark: the performance measure is for generalization outside of training distribution, similar to real-world problems. For example, learn from a social network before 2017, validate on 2018, and compare models on 2019.

0 replies, 7 likes


cs.LG Papers: Open Graph Benchmark: Datasets for Machine Learning on Graphs. Weihua Hu, Matthias Fey, Marinka Zitnik, Yuxiao Dong, Hongyu Ren, Bowen Liu, Michele Catasta, and Jure Leskovec http://arxiv.org/abs/2005.00687

1 replies, 1 likes


Content

Found on May 05 2020 at https://arxiv.org/pdf/2005.00687.pdf

PDF content of a computer science paper: Open Graph Benchmark: Datasets for Machine Learning on Graphs