You may read this chapter in the usual place:
geometricdeeplearning.com/book/ ๐
Any and all feedback / thoughts / questions on the content... words of encouragement for finishing this book (pretty please! ๐) warmly welcomed!
@mmbronstein.bsky.social @joanbruna.bsky.social @taco-cohen.bsky.social
The first draft 'G' chapter of the geometric deep learning book is live! ๐
Alice enters the magical, branchy world of Graphs and GNNs ๐ธ๏ธ (LLMs are there too!)
I've spent 7+ years studying, researching & talking about graphs -- This text is my best attempt at conveying everything i've learnt ๐
...but our model regressed on Quickselect rather than improving it, so I'm indeed curious to explore what you've done differently :)
Cool stuff! Will give it a read.
We also built a 'fully tropical' GNN following similar principles a few years back (from
the perspective of asynchrony invariance), might be a good reference:
arxiv.org/abs/2306.15632
I would say it definitely does! The book does rely on theoretical material but is designed to be a self-contained, deep learning-oriented introduction to those topics.
So if you read the chapters in sequence it _should_ be just fine :)
(would appreciate any feedback if you do try it!)
Read on at
geometricdeeplearning.com/book ๐
Comments, errata reports, and feedback warmly welcomed!
4yrs ago, the Geometric Deep Learning proto-book hit the arXiv ๐
Today, we proudly release *Chapter 4* of the GDL Book! ๐ Wrapping up foundations, exceeding the proto-book page count ๐
Next: deep-dives into the 5 Gs! #โฃ๐๐ธ๏ธ๐๐ชข
@mmbronstein.bsky.social @joanbruna.bsky.social @taco-cohen.bsky.social
The recording of my 'LLMs as GNNs' talk is now public (link in thread) ๐ฌ๐ธ๏ธ
Thank you to everyone in the GLOW community for having me -- I had a fantastic time! ๐
(And I look forward to refining this content even further -- especially as new works come trickling in ๐)
This is happening today! ๐
Join me at the virtual GLOW seminar (5pm CET) for the first public showing of my 'LLMs as GNNs' talk. ๐ฌ๐ธ๏ธ
GLOW is returning on ๐ ๐ฎ๐ฟ๐ฐ๐ต ๐ฎ๐ฒ๐๐ต, ๐ฑ๐ฝ๐บ ๐๐๐ง with a special guest: @petar-v.bsky.social ๐
He will lecture on LLMs as GNNs โ a topic which received quite some attention at our last session.
Specifically, we will learn how Graph ML tools can help understand LLM generalisation
Seminar details + join instructions here:
www.cst.cam.ac.uk/seminars/lis...
Hope to see many of you there!
I'm excited to share that we'll have Ilan Price giving a talk at the University of Cambridge on GenCast -- a state-of-the-art model for probabilistic weather forecasting ๐ฆ๏ธโ๏ธโ
๏ธ
If you're in Cambridge next Wednesday (19 Feb), consider joining us -- it's open to all!
All details (+ Zoom link) below!
It's been great working on this with Alex, @joao.omg.lol and Marc on this over the past couple of months!
We don't believe we solved the problems studied here -- mainly brought them to attention -- and we welcome any and all comments!
Read our work on the arXiv:
arxiv.org/abs/2502.06751
And so we set out to understand _feedforward_ graphs (i.e. graphs w/o back edges) โฉ
Turns out these graphs are rather understudied for how often they are used in practice! ๐ฎ
We work towards a framework to analyse them, which I hope both my GNN and my LLM friends will enjoy ๐
Trying to copy the greats Pietro and @petar-v.bsky.social giving a guest lecture for the GDL class today!
*Round and Round We Go! What makes Rotary Positional Encodings useful?*
by @petar-v.bsky.social et al.
They show RoPE has distinct behavior for different rotation angles - high freq for position, low freq for semantics.
arxiv.org/abs/2410.06205
This summer, a group of elite AI scientists and engineers will come over to Sarajevo ๐ง๐ฆ for #EEML2025 โค๏ธ
We've been working hard behind the scenes to prepare a stellar program for our attendees!
Trust me, you don't want to miss this ๐
Applications open now! (details in post)
Application forms should open early 2025 :)
Exciting news: the NeurIPS Sci4DL Workshop has recognised our softmax paper as a best paper runner-up for its 'Debunking Challenge'! ๐๐งโ๐ฌ
I'm really grateful to Christos and Federico for tirelessly presenting our work throughout the day, and to all attendees who kindly stopped by our talks/posters! ๐
It's official! ๐ I am excited to share that EEML will be coming to Sarajevo ๐ง๐ฆ next summer!
For those not familiar, EEML is the flagship AI summer school for the Eastern European region (and beyond!).
Watch this space! More information on how to apply, lecturers, etc... is coming soon ๐
Today it is time to present our softmax paper to the world! ๐ก๏ธ
If you're at #NeurIPS2024 today, check out our two Workshop spotlights ๐ฆ
๐งโ๐ฌ Sci4DL (by Christos Perivolaropoulos), West Meeting Rooms 205-207, 10:50--11:05am
๐ข System II (by Federico Barbero), West Ballroom B, 9:20--9:30am
A very nice blog from Przemek Pietrzkiewicz, offering thoughts on our recent result in AI for competitive programming ๐
Przemek co-led the Hash Code contest, which we used as the main test-bed to evaluate our approach ๐
Worth a read if you want to understand implications of our work! Link below โฌ๏ธ
If you will be at #NeurIPS2024 @neuripsconf.bsky.social and would like to come see our models in action, come say hi ๐ and check out our demo at the GDM booth!
Wednesday, Dec. 11th @ 9:30-10:00.
Lots of other great things to see as well! Check it out: ๐
deepmind.google/discover/blo...
๐ To apply:
1โฃ Fill out the "Student Researcher 2025" application form for EMEA*:
www.google.com/about/career...
2โฃ Email Larisa Markeeva (the project host) and myself, letting us know you've applied ๐ Also please feel free to contact us if you have any questions!
*this role is onsite in London ๐ฌ๐ง
Our team at Google DeepMind is hiring Student Researchers for 2025!
๐งโ๐ฌ Interested in understanding reasoning capabilities of neural networks from first principles?
๐งโ๐ Currently studying for a BS/MS/PhD?
๐งโ๐ป Have solid engineering and research skills?
๐ We want to hear from you! Details in thread.
I'm excited to share a new paper: "Mastering Board Games by External and Internal Planning with Language Models"
storage.googleapis.com/deepmind-med...
(also soon to be up on Arxiv, once it's been processed there)
Introducing ๐งGenie 2 ๐ง - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents ๐ง .
Just a poster I'm very proud of. Mainly because it's the first time I led a theory-focused paper ๐ฅณ
Coming soon to NeurIPS Workshops near you (two spotlights!!) ๐ฆ
I unfortunately won't be there myself, but Christos and Federico will be around ๐
