Fresh picks from ArXiv
This week on ArXiv: time series recovery, GNN challenge winning solutions, and benchmark for scene graph generation ๐ณ
If I forgot to mention your paper, please shoot me a message and I will update the post.
Applications
* Multivariate Time Series Imputation by Graph Neural Networks
* Temporal-Relational Hypergraph Tri-Attention Networks for Stock Trend Prediction
* Graph Constrained Data Representation Learning for Human Motion Segmentation
* The Graph Neural Networking Challenge: A Worldwide Competition for Education in AI/ML for Networks
* Image Scene Graph Generation (SGG) Benchmark
* Structack: Structure-based Adversarial Attacks on Graph Neural Networks
This week on ArXiv: time series recovery, GNN challenge winning solutions, and benchmark for scene graph generation ๐ณ
If I forgot to mention your paper, please shoot me a message and I will update the post.
Applications
* Multivariate Time Series Imputation by Graph Neural Networks
* Temporal-Relational Hypergraph Tri-Attention Networks for Stock Trend Prediction
* Graph Constrained Data Representation Learning for Human Motion Segmentation
* The Graph Neural Networking Challenge: A Worldwide Competition for Education in AI/ML for Networks
* Image Scene Graph Generation (SGG) Benchmark
* Structack: Structure-based Adversarial Attacks on Graph Neural Networks
Foundations of Graph Neural Networks Course
A new upcoming course by Zak Jost (you may remember his videos on GNNs) on the foundations of GNN which covers such topics as
- Neural Message Passing
- Fourier Transforms, Graph Wavelets and Spectral Convolutions
- Permutation Symmetries
- Representational capacity of GNNs
- Graph fundamentals like the Laplacian and graph isomorphism.
A new upcoming course by Zak Jost (you may remember his videos on GNNs) on the foundations of GNN which covers such topics as
- Neural Message Passing
- Fourier Transforms, Graph Wavelets and Spectral Convolutions
- Permutation Symmetries
- Representational capacity of GNNs
- Graph fundamentals like the Laplacian and graph isomorphism.
Telegram
Graph Machine Learning
GML YouTube Videos
I was pleasantly surprised to see there is YouTube playlist by Zak Jost that covers some aspects of GNNs, including an interview with DeepMind authors for using GNNs for physics.
I was pleasantly surprised to see there is YouTube playlist by Zak Jost that covers some aspects of GNNs, including an interview with DeepMind authors for using GNNs for physics.
Graph Neural Networks: Algorithms and Applications
A great presentation by Jian Tang about GNN basics, training many layers, self-supervised learning and statistical relational learning.
A great presentation by Jian Tang about GNN basics, training many layers, self-supervised learning and statistical relational learning.
Knowledge Graphs in Natural Language Processing @ ACL 2021
A regular update from Michael Galkin on the SOTA applications of KG in the world of words:
Neural Databases & Retrieval
KG-augmented Language Models
KG Embeddings & Link Prediction
Entity Alignment
KG Construction, Entity Linking, Relation Extraction
KGQA: Temporal, Conversational, and AMR.
A regular update from Michael Galkin on the SOTA applications of KG in the world of words:
Neural Databases & Retrieval
KG-augmented Language Models
KG Embeddings & Link Prediction
Entity Alignment
KG Construction, Entity Linking, Relation Extraction
KGQA: Temporal, Conversational, and AMR.
Medium
Knowledge Graphs in Natural Language Processing @ ACL 2021
Your guide to the KG-related NLP research, ACL edition
Essays on Data Science
A great collection of blog posts on machine learning and computer science covering topics such as infinitely wide neural nets, markov models, and graph deep learning.
A great collection of blog posts on machine learning and computer science covering topics such as infinitely wide neural nets, markov models, and graph deep learning.
GDL Course
A course that follows closely the geometric deep learning book. It contains 12 lectures, 2 tutorials, and 4 seminars covering topics such as graphs, sets, grids, groups, geodesics, gauges, and time warping. Videos and slides are available.
A course that follows closely the geometric deep learning book. It contains 12 lectures, 2 tutorials, and 4 seminars covering topics such as graphs, sets, grids, groups, geodesics, gauges, and time warping. Videos and slides are available.
Geometricdeeplearning
GDL Course
Grids, Groups, Graphs, Geodesics, and Gauges
Fresh picks from ArXiv
This week on ArXiv: PDE-inspired GNNs, proves to the conjectures, and a new benchmark for graph completion ๐งต
If I forgot to mention your paper, please shoot me a message and I will update the post.
GNNs
* Distilling Holistic Knowledge with Graph Neural Networks ICCV 2021
* Fully Hyperbolic Graph Convolution Network for Recommendation CIKM 2021
* Jointly Attacking Graph Neural Network and its Explanations
* LEO: Learning Energy-based Models in Graph Optimization
* PDE-GCN: Novel Architectures for Graph Neural Networks Motivated by Partial Differential Equations
* Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training data with Bryan Perozzi
* AdaGNN: A multi-modal latent representation meta-learner for GNNs based on AdaBoosting
Math
* Isomorphisms between random graphs
* Edge Partitions of Complete Geometric Graphs (Part 1)
Survey
* Are Missing Links Predictable? An Inferential Benchmark for Knowledge Graph Completion
* Influence Maximization in Social Networks: A Survey of Behaviour-Aware Methods
This week on ArXiv: PDE-inspired GNNs, proves to the conjectures, and a new benchmark for graph completion ๐งต
If I forgot to mention your paper, please shoot me a message and I will update the post.
GNNs
* Distilling Holistic Knowledge with Graph Neural Networks ICCV 2021
* Fully Hyperbolic Graph Convolution Network for Recommendation CIKM 2021
* Jointly Attacking Graph Neural Network and its Explanations
* LEO: Learning Energy-based Models in Graph Optimization
* PDE-GCN: Novel Architectures for Graph Neural Networks Motivated by Partial Differential Equations
* Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training data with Bryan Perozzi
* AdaGNN: A multi-modal latent representation meta-learner for GNNs based on AdaBoosting
Math
* Isomorphisms between random graphs
* Edge Partitions of Complete Geometric Graphs (Part 1)
Survey
* Are Missing Links Predictable? An Inferential Benchmark for Knowledge Graph Completion
* Influence Maximization in Social Networks: A Survey of Behaviour-Aware Methods
Awesome Efficient Graph Neural Networks
A new awesome repo by Chaitanya K. Joshi with the curated list of must-read papers on efficient Graph Neural Networks and scalable Graph Representation Learning for real-world applications.
A new awesome repo by Chaitanya K. Joshi with the curated list of must-read papers on efficient Graph Neural Networks and scalable Graph Representation Learning for real-world applications.
GitHub
GitHub - chaitjo/efficient-gnns: Code and resources on scalable and efficient Graph Neural Networks
Code and resources on scalable and efficient Graph Neural Networks - chaitjo/efficient-gnns
Book: Designing and Building Enterprise Knowledge Graphs (Synthesis Lectures on Data, Semantics, and Knowledge)
A new book by Ora Lassila and Juan Sequeda that guides on designing and building knowledge graphs from enterprise relational databases in practice. It presents a principled framework centered on mapping patterns to connect relational databases with knowledge graphs, the roles within an organization responsible for the knowledge graph, and the process that combines data and people. The content of this book is applicable to knowledge graphs being built either with property graph or RDF graph technologies.
A new book by Ora Lassila and Juan Sequeda that guides on designing and building knowledge graphs from enterprise relational databases in practice. It presents a principled framework centered on mapping patterns to connect relational databases with knowledge graphs, the roles within an organization responsible for the knowledge graph, and the process that combines data and people. The content of this book is applicable to knowledge graphs being built either with property graph or RDF graph technologies.
Graph Machine Learning research groups: Ian Davidson
I do a series of posts on the groups in graph research, previous post is here. The 33rd is Ian Davidson, a professor at UC Davis, who works in the areas with societal impacts such as neuroscience, intelligent tutoring systems and social networks.
Ian Davidson (~1973)
- Affiliation: UC Davis
- Education: Ph.D. at Monash University in 2000 (advisor: C.S. Wallace)
- h-index 44
- Interests: fairness, clustering, graphical models.
- Awards: best papers at KDD, SIAM, ICDM
I do a series of posts on the groups in graph research, previous post is here. The 33rd is Ian Davidson, a professor at UC Davis, who works in the areas with societal impacts such as neuroscience, intelligent tutoring systems and social networks.
Ian Davidson (~1973)
- Affiliation: UC Davis
- Education: Ph.D. at Monash University in 2000 (advisor: C.S. Wallace)
- h-index 44
- Interests: fairness, clustering, graphical models.
- Awards: best papers at KDD, SIAM, ICDM
Telegram
Graph Machine Learning
Graph Machine Learning research groups: Shuiwang Ji
I do a series of posts on the groups in graph research, previous post is here. The 32nd is Shuiwang Ji, a professor at Texas A&M University. His teams were awarded at OGB-LSC and AI Cures challenges. Heโฆ
I do a series of posts on the groups in graph research, previous post is here. The 32nd is Shuiwang Ji, a professor at Texas A&M University. His teams were awarded at OGB-LSC and AI Cures challenges. Heโฆ
TorchDrug: a powerful and flexible machine learning platform for drug discovery
Jian Tang and his co-workers from MILA open-sourced a new library TorchDrug on drug modeling with machine learning. It includes an easy interface for property prediction, pretrained molecular representations, de-novo molecule design & optimization, knowledge graph reasoning, and more.
Jian Tang and his co-workers from MILA open-sourced a new library TorchDrug on drug modeling with machine learning. It includes an easy interface for property prediction, pretrained molecular representations, de-novo molecule design & optimization, knowledge graph reasoning, and more.
Mila
A team led by Mila researcher Jian Tang launches TorchDrug, an open-source platform for drug discovery - Mila
Mila is a place of collaboration and a meeting point for the main actors of artificial intelligence in Montreal. Our mission is to be a global pole for scientific advances that inspires innovation and the development of AI for the benefit of all.
Fresh picks from ArXiv
This week on ArXiv: GNNs for lidars, complex reasoning in knowledge graphs, and building AI for drawing graphs ๐ค
If I forgot to mention your paper, please shoot me a message and I will update the post.
GNNs
* Generative and Contrastive Self-Supervised Learning for Graph Anomaly Detection
* TabGNN: Multiplex Graph Neural Network for Tabular Data Prediction
* Learning to Match Features with Seeded Graph Matching Network
* EqGNN: Equalized Node Opportunity in Graphs
Applications
* GP-S3Net: Graph-based Panoptic Sparse Semantic Segmentation Network
* Adaptive Graph Convolution for Point Cloud Analysis
* Hyperbolic Hypergraphs for Sequential Recommendation
* Implementation of Sprouts: a graph drawing game
Knowledge graphs
* Fact-Tree Reasoning for N-ary Question Answering over Knowledge Graphs
* UNIQORN: Unified Question Answering over RDF Knowledge Graphs and Natural Language Text
This week on ArXiv: GNNs for lidars, complex reasoning in knowledge graphs, and building AI for drawing graphs ๐ค
If I forgot to mention your paper, please shoot me a message and I will update the post.
GNNs
* Generative and Contrastive Self-Supervised Learning for Graph Anomaly Detection
* TabGNN: Multiplex Graph Neural Network for Tabular Data Prediction
* Learning to Match Features with Seeded Graph Matching Network
* EqGNN: Equalized Node Opportunity in Graphs
Applications
* GP-S3Net: Graph-based Panoptic Sparse Semantic Segmentation Network
* Adaptive Graph Convolution for Point Cloud Analysis
* Hyperbolic Hypergraphs for Sequential Recommendation
* Implementation of Sprouts: a graph drawing game
Knowledge graphs
* Fact-Tree Reasoning for N-ary Question Answering over Knowledge Graphs
* UNIQORN: Unified Question Answering over RDF Knowledge Graphs and Natural Language Text
Video: Graph Neural Networks - a perspective from the ground up
A beautiful video about GNNs aimed at CS undergrads that explains what message passing and node embeddings are and gives a link prediction example.
A beautiful video about GNNs aimed at CS undergrads that explains what message passing and node embeddings are and gives a link prediction example.
YouTube
Graph Neural Networks - a perspective from the ground up
What is a graph, why Graph Neural Networks (GNNs), and what is the underlying math?
Highly recommended videos that I watched many times while making this:
Petar Veliฤkoviฤ's GNN video โ https://youtu.be/8owQBFAHw7E
Michael Bronstein's Geometric Deep Learningโฆ
Highly recommended videos that I watched many times while making this:
Petar Veliฤkoviฤ's GNN video โ https://youtu.be/8owQBFAHw7E
Michael Bronstein's Geometric Deep Learningโฆ
Graph Drawing and Network Visualization 2021
A symposium on graph Drawing and network visualization is a nice niche conference on how to draw graphs efficiently and insightfully. This year it will be organized both online and offline (in Tรผbingen, Germany). Dates are: September 14-17, 2021. Accepted papers can be seen here.
A symposium on graph Drawing and network visualization is a nice niche conference on how to draw graphs efficiently and insightfully. This year it will be organized both online and offline (in Tรผbingen, Germany). Dates are: September 14-17, 2021. Accepted papers can be seen here.
algo.inf.uni-tuebingen.de
Graph Drawing 2021 - GD2021
The 29th International Symposium on Graph Drawing and Network Visualization will be held (hopefully) at Tรผbingen from September 15 to 17, 2021. A pre-conference PhD school is planned for September 13-14, 2021.
Fresh picks from ArXiv
This week on ArXiv: predictions of routing times, benchmarking architecture tricks, and drug repurposing study ๐
If I forgot to mention your paper, please shoot me a message and I will update the post.
Conferences
* Single Node Injection Attack against Graph Neural Networks CIKM 2021
* ETA Prediction with Graph Neural Networks in Google Maps CIKM 2021, with Petar Veliฤkoviฤ
* Tree Decomposed Graph Neural Network CIKM 2021, with Tyler Derr
* DSKReG: Differentiable Sampling on Knowledge Graph for Recommendation with Relational GNN CIKM 2021
* Multiplex Graph Neural Network for Extractive Text Summarization EMNLP 2021
* Demystifying Drug Repurposing Domain Comprehension with Knowledge Graph Embedding IEEE BioCAS 2021
* DC-GNet: Deep Mesh Relation Capturing Graph Convolution Network for 3D Human Shape Reconstruction ACM MM'21
* Visualizing JIT Compiler Graphs GD 2021
GNNs
* Spatio-Temporal Graph Contrastive Learning
Benchmark
* Weisfeiler-Leman in the BAMBOO: Novel AMR Graph Metrics and a Benchmark for AMR Graph Similarity TACL 2021
* Bag of Tricks for Training Deeper Graph Neural Networks: A Comprehensive Benchmark Study
Math
* Smallest graphs with given automorphism group
This week on ArXiv: predictions of routing times, benchmarking architecture tricks, and drug repurposing study ๐
If I forgot to mention your paper, please shoot me a message and I will update the post.
Conferences
* Single Node Injection Attack against Graph Neural Networks CIKM 2021
* ETA Prediction with Graph Neural Networks in Google Maps CIKM 2021, with Petar Veliฤkoviฤ
* Tree Decomposed Graph Neural Network CIKM 2021, with Tyler Derr
* DSKReG: Differentiable Sampling on Knowledge Graph for Recommendation with Relational GNN CIKM 2021
* Multiplex Graph Neural Network for Extractive Text Summarization EMNLP 2021
* Demystifying Drug Repurposing Domain Comprehension with Knowledge Graph Embedding IEEE BioCAS 2021
* DC-GNet: Deep Mesh Relation Capturing Graph Convolution Network for 3D Human Shape Reconstruction ACM MM'21
* Visualizing JIT Compiler Graphs GD 2021
GNNs
* Spatio-Temporal Graph Contrastive Learning
Benchmark
* Weisfeiler-Leman in the BAMBOO: Novel AMR Graph Metrics and a Benchmark for AMR Graph Similarity TACL 2021
* Bag of Tricks for Training Deeper Graph Neural Networks: A Comprehensive Benchmark Study
Math
* Smallest graphs with given automorphism group
GNN Tutorial & Graph Convolution Intuition @ Distill
Distill.pub is a great new resource aimed at re-defining a way we publish papers. Publications on Distill have rich visualizations and hands-on examples that you can tweak right in a browser. Unfortunately, Distill goes on a hiatus.
But, as the last bow, the authors prepared two very cool articles breaking down message passing and graph convolutions:
1. A Gentle Introduction to Graph Neural Networks
2. Understanding Convolutions on Graphs
Something you definitely do not want to miss in September!
Distill.pub is a great new resource aimed at re-defining a way we publish papers. Publications on Distill have rich visualizations and hands-on examples that you can tweak right in a browser. Unfortunately, Distill goes on a hiatus.
But, as the last bow, the authors prepared two very cool articles breaking down message passing and graph convolutions:
1. A Gentle Introduction to Graph Neural Networks
2. Understanding Convolutions on Graphs
Something you definitely do not want to miss in September!
Distill
A Gentle Introduction to Graph Neural Networks
What components are needed for building learning algorithms that leverage the structure and properties of graphs?
Monday Theory: Structural vs Positional Node Representations
In the new slide deck, Bruno Ribeiro (Purdue University) uncovers the nature of two commonly used mechanisms for building node representations. Structural representations are permutation insensitive (like GNNs) whereas positional representations are permutation sensitive (like SVD vectors). Hence, all GRL approaches can be broadly classified into those two families. Takeaway messages:
Message 1: Positional representations of k nodes are to most expressive k-node structural representations as samples of a distribution are to sufficient statistics of the distribution. This is based on the results published in the ICLR'20 paper
Message 2: As soon as you introduce some sort of node IDs you break equivariance but at the same time can predict properties of any subset of nodes (better link prediction). Youโd better aggregate over multiple samples though (from the stats analogy). If you stick to equivariance, you can predict node or graph-level properties but nothing in-between.
In the new slide deck, Bruno Ribeiro (Purdue University) uncovers the nature of two commonly used mechanisms for building node representations. Structural representations are permutation insensitive (like GNNs) whereas positional representations are permutation sensitive (like SVD vectors). Hence, all GRL approaches can be broadly classified into those two families. Takeaway messages:
Message 1: Positional representations of k nodes are to most expressive k-node structural representations as samples of a distribution are to sufficient statistics of the distribution. This is based on the results published in the ICLR'20 paper
Message 2: As soon as you introduce some sort of node IDs you break equivariance but at the same time can predict properties of any subset of nodes (better link prediction). Youโd better aggregate over multiple samples though (from the stats analogy). If you stick to equivariance, you can predict node or graph-level properties but nothing in-between.
Fresh picks from ArXiv
This week on ArXiv: improving robustness by resampling a graph, learning better scenes, and new homophily definitions ๐ค
If I forgot to mention your paper, please shoot me a message and I will update the post.
GNNs
* Training Graph Neural Networks by Graphon Estimation
* Learning to Generate Scene Graph from Natural Language Supervision ICCV 2021
* Pointspectrum: Equivariance Meets Laplacian Filtering for Graph Representation Learning
* Sparsifying the Update Step in Graph Neural Networks
* Adaptive Label Smoothing To Regularize Large-Scale Graph Training
Math
* How likely is a random graph shift-enabled?
* The Popularity-Homophily Index: A new way to measure Homophily in Directed Graphs
This week on ArXiv: improving robustness by resampling a graph, learning better scenes, and new homophily definitions ๐ค
If I forgot to mention your paper, please shoot me a message and I will update the post.
GNNs
* Training Graph Neural Networks by Graphon Estimation
* Learning to Generate Scene Graph from Natural Language Supervision ICCV 2021
* Pointspectrum: Equivariance Meets Laplacian Filtering for Graph Representation Learning
* Sparsifying the Update Step in Graph Neural Networks
* Adaptive Label Smoothing To Regularize Large-Scale Graph Training
Math
* How likely is a random graph shift-enabled?
* The Popularity-Homophily Index: A new way to measure Homophily in Directed Graphs
The Learning on Graphs and Geometry Reading Group
A reading group organized by Hannes Stรคrk with supervision from Pietro Liรฒ at Cambridge. Includes really interesting fresh papers on graphs. Every Tuesday at 5pm CEST.
A reading group organized by Hannes Stรคrk with supervision from Pietro Liรฒ at Cambridge. Includes really interesting fresh papers on graphs. Every Tuesday at 5pm CEST.
Researcher Positions at Dimitri's Ognibene's Lab
Two positions for post doc/researchers are available at Milano Bicocca University under Dimitri's Ognibene supervision. 2 years contract, based in Milan (possibility to remote working). For application contact: Dimitri Ognibene [email protected]. Description is below:
Do social media harm teenagers and our society?
Can we make them safer?
We will use the state of the art in graph neural networks, reinforcement learning, nlp, cv, and machine learning in general to improve our understanding of social media dynamics, and help our society by supporting and teaching young people tackle hate speech and fake news in social media.
Two positions for post doc/researchers are available at Milano Bicocca University under Dimitri's Ognibene supervision. 2 years contract, based in Milan (possibility to remote working). For application contact: Dimitri Ognibene [email protected]. Description is below:
Do social media harm teenagers and our society?
Can we make them safer?
We will use the state of the art in graph neural networks, reinforcement learning, nlp, cv, and machine learning in general to improve our understanding of social media dynamics, and help our society by supporting and teaching young people tackle hate speech and fake news in social media.
Google
Dimitri Ognibene's Homepage - Jobs
Currently Open Positions
Graph ML in Industry Workshop
When I wrote top applications of GNNs at the beginning of this year, I had a feeling that graph ML community is mature enough to start being used in industrial companies. Nine months ahead we decided to gather researchers, engineers, and industry professionals to talk about applications of graphs in the companies. Please, join us on 23rd Sept, 17-00 Paris time (free, online, ~3 hours) by registering at the link.
When I wrote top applications of GNNs at the beginning of this year, I had a feeling that graph ML community is mature enough to start being used in industrial companies. Nine months ahead we decided to gather researchers, engineers, and industry professionals to talk about applications of graphs in the companies. Please, join us on 23rd Sept, 17-00 Paris time (free, online, ~3 hours) by registering at the link.
Google
Graph Machine Learning in Industry
Criteo AI Lab is excited to be presenting Graph Machine Learning in Industry. Please join us on Thursday, September 23rd, at 17:00 Paris time. This page will be updated with video links after the workshop.