Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) 
https://decompwlj.com/3DgraphGen/Natural_numbers.html
#Graphs
18 posts8 participants1 post today
Replied in thread
Now this animation is available for the 1000 sequences decomposed on my website.
Accessible from the 3Dgraph, 2Dgraph500terms and 2dgraphs pages https://decompwlj.com
A little more work on axis sizing and controls.
decompwlj.comDecomposition into weight × level + jump - 3D graphs - 2D graphs - First 500 terms - Rémi EismannDecomposition into weight × level + jump with 3D graphs (WebGL three.js), 2D graphs and first 500 terms. This decomposition is an extension of the fundamental theorem of arithmetic and a new way to see the numbers. 1000 sequences decomposed (natural numbers, prime numbers, triangular numbers, composites...). Rémi Eismann
Replied in thread
4: The palindromes in base 10 (A002113) https://decompwlj.com/3DgraphGen/Palindromes.html
Continued thread
3: The triangular numbers (A000217) https://decompwlj.com/3DgraphGen/Triangular_numbers.html
Continued thread
2: The prime numbers (A000040) https://decompwlj.com/3DgraphGen/Prime_numbers.html
Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) https://decompwlj.com/3DgraphGen/Natural_numbers.html
Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) https://decompwlj.com/3DgraphGen/Natural_numbers.html
And as an added bonus: Dileep George, one of the authors of the paper, just shared a #JupyterNotebook demo 
. You can explore the #CSCG model, visualize #PlaceFields, and inspect the learned #latent #graphs 
Just try it out, it's great fun 
colab.research.google.comGoogle Colab
This paper by Raju et al. proposes a unified model – “clone‑structured causal #graphs” (#CSCG) – for #hippocampal #SpatialCoding. It suggests that #SpatialMaps arise from #learning #latent higher‑order sequences rather than representing #EuclideanSpace directly. The model elegantly explains phenomena like #PlaceFields, #SplitterCells, #contextual #remapping, and predicts when #PlaceFieldMapping may mislead.
Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) https://decompwlj.com/3DgraphGen/Natural_numbers.html
Continued thread
Now this animation is available for the 1000 sequences decomposed on my website.
Accessible from the 3Dgraph, 2Dgraph500terms and 2dgraphs pages https://decompwlj.com
A little more work on axis sizing and controls.
decompwlj.comDecomposition into weight × level + jump - 3D graphs - 2D graphs - First 500 terms - Rémi EismannDecomposition into weight × level + jump with 3D graphs (WebGL three.js), 2D graphs and first 500 terms. This decomposition is an extension of the fundamental theorem of arithmetic and a new way to see the numbers. 1000 sequences decomposed (natural numbers, prime numbers, triangular numbers, composites...). Rémi Eismann
Continued thread
4: The palindromes in base 10 (A002113) https://decompwlj.com/3DgraphGen/Palindromes.html
Continued thread
3: The triangular numbers (A000217) https://decompwlj.com/3DgraphGen/Triangular_numbers.html
Continued thread
2: The prime numbers (A000040) https://decompwlj.com/3DgraphGen/Prime_numbers.html
Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) https://decompwlj.com/3DgraphGen/Natural_numbers.html
Continued thread
I have covered the results of elections for the US House of Representatives.
https://jasonbeets.blogspot.com/2025/01/house-of-representatives.html
8/x
jasonbeets.blogspot.comRepublicans retain House control In 2024, Republicans retained control of the US House of Representatives. Republicans won 220 seats, and Democrats won 215 seats. ...
Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) https://decompwlj.com/3DgraphGen/Natural_numbers.html
Now out in #TMLR:
GRAPES: Learning to Sample Graphs for Scalable Graph Neural Networks
There's lots of work on sampling subgraphs for GNNs, but relatively little on making this sampling process _adaptive_. That is, learning to select the data from the graph that is relevant for your task.
We introduce an RL-based and a GFLowNet-based sampler and show that the approach performs well on heterophilic graphs.
All those who couldn't join us in Natal for the International School and Workshop in Complex Networks Beyond Pairwise Interactions can still access all the lectures and talks, courtesy of the International Institute of Physics! 
- Fundamental definitions and properties
- Dynamical processes
- Synchronisation
- Control
- Belief propagation
- Community detection
and more!
Please boost to reach the largest possible audience.
https://www.youtube.com/playlist?list=PLqTLz9G2bGs2-ToEsAuUg3u6ZASY1UbAk
Generation of four sequences decomposed into weight × level + jump (log(weight), log(level), log(jump)) - three.js animation:
1: The natural numbers (A000027) https://decompwlj.com/3DgraphGen/Natural_numbers.html