Lightnews — Scholar-powered news

Christopher Morris @chrsmrrs.bsky.social · Jul 24

This will be great newgraphperspectives.com.

New Perspectives in Advancing Graph Machine Learning

NeurIPS 2025 Workshop

newgraphperspectives.com

3

Christopher Morris @chrsmrrs.bsky.social · Jul 23

If you are in the Bay Area, consider attending our workshop, "Graph Learning Meets TCS," at the Simons Institute (simons.berkeley.edu/workshops/gr...).

Graph Learning Meets Theoretical Computer Science

Graph learning is a branch of machine learning focusing on developing and studying methods to make predictions for vertices, subgraphs, or entire graphs. The field of graph learning has already reveal...

simons.berkeley.edu

1 1 4

Christopher Morris @chrsmrrs.bsky.social · May 5

An interesting paper by Eran Rosenbluth and Martin Grohe, www.arxiv.org/abs/2505.00291, shows a connection between polynomial-time algorithms and recurrent GNNs.

Repetition Makes Perfect: Recurrent Sum-GNNs Match Message Passing Limit

We provide first tight bounds for the expressivity of Recurrent Graph Neural Networks (recurrent GNNs) with finite-precision parameters. We prove that recurrent GNNs, with sum aggregation and ReLU act...

www.arxiv.org

Reposted by Christopher Morris

NeurIPS Conference @neuripsconf.bsky.social · May 2

Responsible reviewing initiatives for NeurIPS 2025 - read more about changes to reviewing that that will safeguard reviewing quality and timeline in our blog post below:
blog.neurips.cc/2025/05/02/r...

Responsible Reviewing Initiative for NeurIPS 2025 – NeurIPS Blog

Communications Chairs 2025 2021 Conference

blog.neurips.cc

1 6 27

Reposted by Christopher Morris

Michael Bronstein @mmbronstein.bsky.social · Apr 29

A reminder that @aithyra.bsky.social has an open call for starting PIs. We offer some of the best conditions in Europe rivalling MP and ETHZ

www.oeaw.ac.at/aithyra/news...

1 7 32

Reposted by Christopher Morris

Kimon Fountoulakis @kfountou.bsky.social · Mar 9

Computational Capability and Efficiency of Neural Networks: A Repository of Papers

I compiled a list of theoretical papers related to the computational capabilities of Transformers, recurrent networks, feedforward networks, and graph neural networks.

Link: github.com/opallab/neur...

2 1 10

Reposted by Christopher Morris

logconference.bsky.social @logconference.bsky.social · Apr 2

🚨 We’re recruiting organizers for LoG 2025!

Join us in shaping the next Learning on Graphs conference at UCLA this fall.

Passionate about graph ML & building community? Apply by Apr 11 (AOE) 🌍

📋 Form: docs.google.com/forms/d/e/1F...

🏠LoG website: logconference.org

This link will take you to a page that’s not on LinkedIn

lnkd.in

4 5

Christopher Morris @chrsmrrs.bsky.social · Apr 2

www.dfg.de/en/news/news...

Emmy Noether Groups “Methods in Artificial Intelligence”

www.dfg.de

Christopher Morris @chrsmrrs.bsky.social · Mar 24

Thanks, Leo!

1

Reposted by Christopher Morris

Leonardo Cotta @cottascience.bsky.social · Mar 24

This is my favourite "graph paper" of the last 1 or 2 years. We also need to start including non-NN baselines, e.g. fingerprints+catboost ---if the goal is real-world impact and not getting it published asap. I also recommend following @wpwalters.bsky.social's blog.
arxiv.org/abs/2502.14546

Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks

While machine learning on graphs has demonstrated promise in drug design and molecular property prediction, significant benchmarking challenges hinder its further progress and relevance. Current bench...

arxiv.org

1 1 7

Christopher Morris @chrsmrrs.bsky.social · Mar 21

Joint work with Antonis Vasileiou, Ben Finkelshtein, Ron Levie, and Floris Geerts.

Christopher Morris @chrsmrrs.bsky.social · Mar 21

In arxiv.org/abs/2412.07106, we leverage modern graph similarity—capturing the fine-grained geometry of MPNNs' feature space—to derive generalization bounds for MPNNs. Our theory is broad, covering various aggregation and loss functions.

1 2 2

Christopher Morris @chrsmrrs.bsky.social · Mar 21

Want to know about the current understanding of the generalization abilities of GNNs? Please have a look at our survey paper arxiv.org/abs/2503.15650.

Joint work with Antonis Vasileiou, Stefanie Jegelka, and Ron Levie.

2 4

Reposted by Christopher Morris

Michael Bronstein @mmbronstein.bsky.social · Mar 17

Aithyra is hiring! We are looking for multiple starting PIs in life sciences and AI/ML. We offer highly competitive packages and salaries among the highest in Europe.

www.oeaw.ac.at/aithyra/news...

Search for LIFE SCIENCE and AI/ML Starting Principal Investigators (m/f/d) at AITHYRA, Vienna

If you are an ambitious early-career scientist ready to propel your research to new heights, we invite you to be part of AITHYRA’s founding generation of Principal Investigators. Together, we will sha...

www.oeaw.ac.at

2 13 39

Christopher Morris @chrsmrrs.bsky.social · Mar 4

Very interesting work arxiv.org/pdf/2503.01431 from @bifold.berlin showing that @luismueller.bsky.social's
edge transformer arxiv.org/abs/2401.10119 achieves SOTA performance on molecular dynamics.

Weisfeiler and Leman strike again. 💪

arxiv.org

2 6

Reposted by Christopher Morris

BIFOLD Berlin Institute for the Foundations of Learning and Data @bifold.berlin · Jan 6

Join BIFOLD as a doctoral student in our Graduate School. Apply until February 03, 2025.

Job ID: www.jobs.tu-berlin.de/en/job-posti...

#jobvacancy #jobalert #postdocposition #postdocfellowship
#ScienceJobs #AI #research #Berlin @tuberlin.bsky.social

The Graduate School educates students for the rising worldwide demand for specialists with expertise in Data Management and Machine Learning. With a strong research focus, we are committed to training the next generation of curious and innovative data science professionals - individuals who challenge assumptions, seek answers to critical questions, and uphold ethical standards in their work.

1 4 8

Reposted by Christopher Morris

Andreas @andreas.de · Feb 1

Der Informatiker @kerstingaiml.bsky.social (TU Darmstadt / @dfki.bsky.social) ordnet im FAZ-Podcast Digitalwirtschaft #DeepSeek ein und welche Chancen sich für Europa öffnen. overcast.fm/+AANfZj8LS4w

Deepseek rüttelt die KI-Welt auf – eine Chance gerade für Deutschland — F.A.Z. Digitalwirtschaft

Haben die Chinesen in der Künstlichen Intelligenz geschafft, wovon Amerikaner und Europäer bislang nur träumen? Haben die Tüftler eines Start-ups aus Fernost die führenden Internetkonzerne des Westens...

overcast.fm

3 5

Reposted by Christopher Morris

icmlconf.bsky.social @icmlconf.bsky.social · Jan 5

ICML 2025 Call For Tutorials

icml.cc/Conferences/...

ICML 2025 Call For Tutorials

icml.cc

3 4

Christopher Morris @chrsmrrs.bsky.social · Jan 2

Gott sei Dank! :P

1

Reposted by Christopher Morris

Leonardo Cotta @cottascience.bsky.social · Dec 22

very cool observations about using smiles/graphs vs fingerprints. TDLR; fingerprints only capture certain properties marginally, and their combinations can often give rise to something new/different.
www.deepmedchem.com/articles/wha...

What Can Neural Network Embeddings Do That Fingerprints Can’t?

www.deepmedchem.com

1 5

Reposted by Christopher Morris

BIFOLD Berlin Institute for the Foundations of Learning and Data @bifold.berlin · Dec 20

go.bsky.app/HN2MTzp

1 1

Reposted by Christopher Morris

Fabrizio Frasca @ffabffrasca.bsky.social · Dec 3

🌟 GLOW is coming back in December with amazing speakers: Emily Jin and @joshsouthern.bsky.social !

🗓️ Dec 18th @ 17 CET on Zoom, don't miss that!

🌐 Find more here: sites.google.com/view/graph-l...

1 5 14

Reposted by Christopher Morris

Leonardo Cotta @cottascience.bsky.social · Dec 11

If you're interested in {causality, language, healthcare}, stop by!
Thursday 11am - 2pm
West Ballroom A-D #5110

Rahul G. Krishnan @rahulgk.bsky.social · Dec 11

@nikitadhawan.bsky.social developed NATURAL (www.cs.toronto.edu/~nikita/natu...) with @cottascience.bsky.social , Karen & @cmaddis.bsky.social. Its an end-to-end pipeline that starts from raw-text data and ends with a causal (**) effect associated with an intervention.

(**) conditions apply
🧵(6/7)

NATURAL

www.cs.toronto.edu

1 3

Christopher Morris @chrsmrrs.bsky.social · Dec 8

🤫😎

1

Christopher Morris @chrsmrrs.bsky.social · Dec 8

Me neither. 😅

1 3