Harry de los Ríos

Network Science | Complex Systems

[NetScience @net_science]

Opinion dynamics: Statistical physics and beyond https://t.co/JRQYWg57Qb

[Neferchitty @Neferchitty]

El sistema GOLLuM (Gaussian process Optimized LLMs) modera el exceso de confianza de los LLM para que aprendan a partir de resultados de incertidumbre

https://t.co/kItHq0Xft0

@hsocasnavarro @pcoffeebreak

[NetScience @net_science]

Criticality and universality in network dismantling https://t.co/gN4259i59u

[Didier 'Dirac's ghost' Gaulin @DiracGhost]

If you are a mathematician or a physicist with an interest in geometry (as you should!), you need to read this fantastic text (70+ pages) titled ''Erlangen Programme at Large: An Overview'' by the excellent Vladimir Kisil.

It's a wonderful introduction to the Erlangen Programme, full of visualizations and examples, in geometry, but also, in other domains of mathematics, and physics as well.

Requirements: Geometry, complex analysis, abstract algebra/group theory.

🔗👇👇👇👇

[Nirosha J. Murugan @niroshajmurugan]

You emit light.

Expanding biology beyond molecules allows us to consider another layer of life: how energy, including light, may reflect and perhaps help coordinate biological activity.

Thank you to @JoMarchant and @Nature for featuring our work alongside other leaders in the field, and for opening a broader conversation about what might be possible, the tools we must build, and the next generation of scientists we must train.

Now, we have the opportunity to determine what the light of life is telling us.

[PNASNews @PNASNews]

One of the most-viewed PNAS articles in the last week is “Why human societies adopt rigid moral rules: The efficiency–robustness trade-off.” Explore the article here: https://t.co/sHfvHujYJX

For more trending articles, visit https://t.co/uz5fLfOeAR.

[論計舎 @ronkeisha_info]

「情報とは何か」を論理で定義できるか。真理性を推論可能性に置き換え、「inferon」という情報の基本単位を証明論的意味論で定式化する試み。Dretske の情報論を証明論に接続する挑戦的なアプローチ。

https://t.co/TRpucnxri7
#証明論 #生成AI
https://t.co/dySKh3RPEv

[Ethan Siegel @StartsWithABang]

The physical Universe isn’t made of pure mathematics alone

Mathematics is great, as a tool, for describing the Universe.

But our Universe isn't made of pure mathematics alone.

To transform it into physics, something more is required.

https://t.co/tYx3HQBiy5

[José R. Berrendero @JRBerrendero]

He subido al blog el material que he utilizado en la asignatura Estadística Aplicada (Grado en Bioquímica de la UAM).

https://t.co/DH5gklHBPU

[Pengrui Han (Barry) @pengrui_han]

The human brain is strikingly modular, with distinct networks for language, formal reasoning, social reasoning, and physical reasoning. Is this a fundamental principle of how intelligent systems are built, or an accident of biological evolution?

In our latest preprint, we find that a similar modular organization emerges in Large Language Models, another class of intelligent system.

Brains and LLMs are shaped by entirely different kinds of optimization (biological evolution vs. gradient descent). That they arrive at the same modular design anyway suggests modularity may be a fundamental property of intelligent systems.

🌐 Web: https://t.co/ZKrnTSSuSf
📄 Paper: https://t.co/ZibBXz3PUy
💻 Code & data: https://t.co/uBo5iOYNjy

Using circuit analyses across 46 tasks spanning four cognitive domains, we find:

1️⃣ Tasks that draw on the same network in humans recruit overlapping units in LLMs, while tasks drawing on different networks recruit distinct units.

2️⃣ These units are causally linked to model behavior. Ablating the units critical for one domain impairs performance in that domain (−26% accuracy) but barely touches the others (−2.5%).

This project has been in the works for a while :) Huge thanks to my advisors @jacobandreas @ev_fedorenko @devarda_a, and to @Nancy_Kanwisher for valuable conceptual input and feedback throughout. #MIT

[Rosana Ferrero 📈📊🙌 @RosanaFerrero]

¿Cómo evalúas y diagnosticas tu modelo de regresión?
Los modelos de regresión están basados en ciertos supuestos que deben cumplirse para que las estimaciones sean válidas y confiables.
🧪Diagnóstico gráfico de residuos: lo que todo analista debe revisar👇🧵

#RStats #DataScience

[Didier 'Dirac's ghost' Gaulin @DiracGhost]

Influenced by the book 'Visual Differential Geometry and Forms' , I will recommend to you on this Sunday, the primer 'A Visual Introduction to Curved Geometry for Physicists' by Urbanski, a visual introduction to the basics of differential geometry, publicly available on arXiv.

Few requirements are necessary, other than (very) basic vector calculus and special relativity.

🔗👇

[NetScience @net_science]

Community detection by the normalized Ricci flow with the optimization of the information entropy https://t.co/C4j0JV9zhA

[NetScience @net_science]

Information theory for hypergraph similarity | Science Advances https://t.co/09a7pEi5SU

[Universitat Rovira i Virgili @universitatURV]

🔝Els investigadors de la #URV Roger Guimerà, Clara Granell, Francesc Giralt, Sergio Gómez i Àlex Arenas signen dos articles que han estat seleccionats entre els més influents dels darrers 30 anys en l’àmbit de la física estadística. Enhorabona!

👉https://t.co/MNVw4euG5J

[Albert-László Barabási @barabasi]

Does teleportation erase the laws of human mobility? In our new paper with @kishore__vasan & @MartonKarsai, we follow 163,770 Decentraland users and their NFT trajectories. The answer: virtual worlds carry the fingerprints of human mobility.
PDF here: https://t.co/AHR1woNJCO

[Raghu Mahajan @raghumahajan]

It is really fascinating to read the commentary by mathematicians in this arXiv posting accompanying the recent announcement by OpenAI about the unit distance problem:

https://t.co/ct4cX60H5e

[SEES Lab @sees_lab]

Can we design generative #AI models capable of creating new molecules, much in the same way that other generative models generate text or images? In a new paper in @NatMachIntell, we introduce #CoCoGraph, which does just that

https://t.co/SSDmPBDZ4e
https://t.co/BzIcmFsooL

[NetScience @net_science]

Predictability of complex networks https://t.co/iFgZT4uIqO

[Quanta Magazine @QuantaMagazine]

Category theory is a mathematical way to formalize the relationships between objects. It could provide a powerful method for modeling complex, interconnected systems. Read @nattyover's new Qualia column: https://t.co/Kr2BxCHvMK