Ben Finkelshtein
@benfinkelshtein
PhD student at @UniofOxford site: https://benfinkelshtein.github.io/
Fun times at ICML. Graph learning dinner, position poster gang, theory, and graph learning hike. :)
At ICML 🇨🇦 presenting the spicy 🌶️ Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks 📍 East Hall A-B #E-604, Thu Also, @antvas98 will be presenting "Covered Forest" — glad to have played a part in this one! 📍 #E-2908, Thu DM to chat graph(+foundation models)


MSR Redmond. Just in time for the 4th 🇺🇸🎆 #MSRIntern

📣 Our spicy ICML 2025 position paper: “Graph Learning Will Lose Relevance Due To Poor Benchmarks”. Graph learning is less trendy in the ML world than it was in 2020-2022. We believe the problem is in poor benchmarks that hold the field back - and suggest ways to fix it! 🧵1/10
Good weather makes meetings inside a gloomy office… unattractive. We’d rather go to University Parks.
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.
* Graph Learning Will Lose Relevance Due To Poor Benchmarks * Maya Bechler-Speicher, @benfinkelshtein, @ffabffrasca, Luis Muller, et al. Inc. @mmbronstein @phanein @michael_galkin @chrsmrrs Wow we should be discussing / acting on this!
I wish @CVPRConf had the guts to make a similar statement about a similar shameful episode this summer. Instead, they washed their hands.
Come check out Learning on Large Graphs using Intersecting Communities! (With a “hint” of Game of Throne references) @NeurIPS 📌East Exhibit Hall A-C #3001, Session 3

If you're a Game of Thrones ⚔️ or Graph Learning fan, I'll present "Learning on Large Graphs using Intersecting Communities" @NeurIPS 📌East Exhibit Hall A-C #3001, Session 3 ⏲️Thu 12 Dec 11-14 Reach out to chat about Geometric DL! Thanks @ismaililkanc @mmbronstein @levie_ron

I appreciate the solidarity but I have to emphasize that never in my lifetime have I seen such appalling mob-incited unbridled Jew hatred being endorsed and emboldened by one of the most respected institutions in academia. The entire debate was a type of Dreyfus trial and it was…
The treatment of @YosephHaddad is a damning indictment on @UniofOxford and hideous reminder that Jew-hate has been completely emboldened and renormalised. Solidarity with Yoseph and @emilykschrader.
Thank you to everyone who came to the @LogConference meetup in Oxford yesterday! Also, shoutout to @jacobbamberger @TeoReu @epomqo @mmbronstein for co-organizing and making this possible! :-)
Learning on Large Graphs using Intersecting Communities is accepted to @NeurIPSConf #neurips2024! An efficient algorithm with linear memory/time following our constructive Weak Graph Regularity Lemma Thanks to @ismaililkanc @mmbronstein @levie_ron Paper:arxiv.org/abs/2405.20724
A heartfelt thank you to @chrsmrrs for hosting me in his group at @RWTH. A month full of engaging discussions, an exciting research collaboration and more rain than I ever thought possible. IMG: Chris kindly displaying a post-CrossFit pic of Antonis & me to his entire class😶🌫️

Cooperative message passing @benfinkelshtein @hxyscott
I was deeply offended by a slide in a recent talk at #CVPR2024 that falsely accused my country of genocide. Such baseless political statements have no place in our scientific community. Let's keep our focus on advancing science and leave politics at the door. @CVPR
Grateful for a delightful dinner with my amazing supervisors @mmbronstein @ismaililkanc #PostNeurIPSGoals

Speak no GNNs 🙊, See no GNNs 🙈... but definitely READ 📖 about CoGNNs, a new message-passing paradigm, which has been accepted to #ICML2024! 🎉 Special thanks to @hxyscott, @mmbronstein and @ismaililkanc. Paper: lnkd.in/ejqYucxi
bit.ly/3PLHs5s **Cooperative GNNs** with @benfinkelshtein @hxyscott @ismaililkanc In message-passing GNNs, each node is updated based on messages from its neighborhood. In Co-GNNs, every node can choose to either ‘listen’, ‘broadcast’, ‘listen&broadcast’, or ‘isolate’
Co-GNN paper is now accepted to @icmlconf See you in Vienna!
bit.ly/3PLHs5s **Cooperative GNNs** with @benfinkelshtein @hxyscott @ismaililkanc In message-passing GNNs, each node is updated based on messages from its neighborhood. In Co-GNNs, every node can choose to either ‘listen’, ‘broadcast’, ‘listen&broadcast’, or ‘isolate’