Unlock the Black Box by Interpreting Graph Convolutional Networks via Additive Decomposition

https://openreview.net/forum?id=sroF8hhbzW

#gnn #subgraph #interpretability

LightGlue: Local Feature Matching at Light Speed

Improving SuperGlue with changes to transformer (GNN matching). Iterative design gives speed boost.

My summary on HFPapers: https://huggingface.co/papers/2306.13643#64ba93f75e13c0d8659b4ce8

Links: [PapersWithCode](https://paperswithcode.com/paper/lightglue-local-feature-matching-at-light), [GitHub](https://github.com/cvg/lightglue), [arxiv](https://arxiv.org/abs/2306.13643)

#lfm #arxiv #PaperThread #NewPaper #gnn

Am Wochenende geht es mit #1000TanksFürNürnberg los. Eine schöne Ergänzung zum #Bardentreffen in #Nürnberg!
#sdgsgolocal #GNN

https://sdgs-go-local.bluepingu.de/pressemitteilung-1000tanksfuernuernberg-popup-event-und-designwettbewerb/

#Introduction I'm a 3rd year PhD student focusing on brining #haemodynamics into the clinic in the context of #cardiovascular research. To do this, I use novel graph neural network (#gnn) architectures to provide on-the-fly uncertainty quantification for clinical metrics. I regularly use #python, #r, #fortran and #sql. I plan to post mainly about interesting papers, repositories and my own work.

Graph-based Multi-ODE Neural Networks for Spatio-Temporal Traffic Forecasting

Zibo Liu, Parshin Shojaee, Chandan K. Reddy

Action editor: Ivan Oseledets.

https://openreview.net/forum?id=Oq5XKRVYpQ

#traffic #transportation #gnn

FASTRAIN-GNN: Fast and Accurate Self-Training for Graph Neural Networks

Amrit Nagarajan, Anand Raghunathan

https://openreview.net/forum?id=1IYJfwJtjQ

#supervised #trained #gnn

Releasing Graph Neural Networks with Differential Privacy Guarantees

https://openreview.net/forum?id=wk8oXR0kFA

#privacy #privgnn #gnn

Graph Neural Networks Designed for Different Graph Types: A Survey

Josephine Thomas, Alice Moallemy-Oureh, Silvia Beddar-Wiesing, Clara Holzhüter

https://openreview.net/forum?id=h4BYtZ79uy

#graphs #graph #gnn

Fast&Fair: Training Acceleration and Bias Mitigation for GNNs

https://openreview.net/forum?id=nOk4XEB7Ke

#normalization #gnn #bias

FASTRAIN-GNN: Fast and Accurate Self-Training for Graph Neural Networks

https://openreview.net/forum?id=1IYJfwJtjQ

#trained #supervised #gnn

Graphcore's partnership with PNNL (part of the US Department of Energy) unlocks dramatically faster 3D molecular modelling using SchNet Graph Neural Network #GNN https://t.co/Lv2dpaKUMK

Node-Level Differentially Private Graph Neural Networks

https://openreview.net/forum?id=u1hp5xGiYW

#privacy #private #gnn

🧙‍♂️ "Show me your NFT and I tell you how it will perform" ! 🧙‍♀️

💥 Check out our postprint of "Multimodal representation learning for NFT selling price prediction", just accepted at #TheWebConf2023 🎉

📄 arxiv.org/abs/2302.01676

📝 w/ @starquake and @andreatagarelli

#TheWebConf #WWW2023 #NFT #Web3 #Blockchain #Metaverse #AI #NLP #NetworkScience #MachineLearning #DeepLearning #Transformers #ComputerVision #GNN

@webscience @economics
@networkscience
@complexsystems
@computationalsocialscience

Postdoc position in graph reasoning for narrative
https://www.rug.nl/about-ug/work-with-us/job-opportunities/?details=00347-02S0009QSP
Deadline: 19 Feb

Tasks: ontology learning, semantic relation identification, extracting structured information from narrative texts. #NLProc #DH #GNN #wikidata #fanfic #job #academic
@gronlp

Do Transformer Really Perform Badly for Graph Representation Learning?

https://www.microsoft.com/en-us/research/project/graphormer/

https://arxiv.org/abs/2106.05234

https://github.com/Microsoft/Graphormer

#GNN, #AI, #ML, #DeepLearning, #Graphs

Ok, finally found some time to play with #ModelAngelo
@SjorsScheres I’m assuming that for homooligomers I just repeat the seq in the fasta file? #cryoEM #GNN

You can sample nodes for scalable #GNN #training. But how do you do #scalable #inference?

In our latest paper (Oral #LogConference
) we introduce influence-based mini-batching (#IBMB) for both fast inference and training, achieving up to 130x and 17x speedups, respectively!

1/8 in 🧵

This seems like a promising approach: a bit of physics informed and has flavor of DDA.
https://pubs.acs.org/doi/full/10.1021/acsphotonics.2c01019
'a Graph Neural Networks (GNN) architecture which learns to model #electromagnetic scattering, can be applied to metasurfaces of arbitrary sizes. Most importantly, it takes into account the coupling between scatterers. Using this approach, near-fields of #metasurfaces with dimensions spanning hundreds of times the wavelength can be obtained in seconds.' #photonics #GNN #MachineLearing

Presenting a hybrid algorithm able to find groups of users who can influence as many people as possible in: “Boosting a Genetic Algorithm with Graph Neural Networks for Multi-Hop Influence Maximization in Social Networks” by Camilo Chacón Sartori, Christian Blum. Proceedings of the 17th Conference on Computer Science and Intelligence Systems, appeared in: M. Ganzha, L. Maciaszek, M. Paprzycki, D. Ślęzak (eds); ACSIS, Vol. 30, pages 363–371 (2022)
#GNN #BRKGA
Open Access: https://lnkd.in/dn5QZGqf

Towards Geometric Deep Learning, article series by Michael Bronstein, Oxford.

#DeepLearning #GeometricDeepLearning
#NeuralNetworks #GNN

https://towardsdatascience.com/towards-geometric-deep-learning-i-on-the-shoulders-of-giants-726c205860f5