Toots for maxpool@mathstodon.xyz account

Written by ma𝕏pool on 2025-01-31 at 11:26

Unjournal: Public journal-independent evaluation of impactful research https://www.unjournal.org/

The Unjournal (unjournal.org) is a nonprofit that commissions (and pays) experts to publicly evaluate and rate publicly-hosted research. It prioritizes globally-impactful work in economics and quantitative social science. It doesn’t charge for anything and it’s all open-access: you can see the evaluation output at https://unjournal.pubpub.org/

[#]science #PeerReviewing #metascience

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-20 at 15:26

Ten Percent Of U.S. High School Students Graduating Without Basic Object Permanence Skills

https://www.youtube.com/watch?v=ssjokgx0pUQ

[#]funny

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-19 at 13:57

Exclusive: German ambassador warns of Trump plan to redefine constitutional order, document shows https://www.reuters.com/world/us/german-ambassador-warns-trump-plan-redefine-constitutional-order-document-shows-2025-01-18/

Germany's ambassador to the United States has warned that the incoming Trump administration will rob U.S. law enforcement and the media of their independence and hand big tech companies "co-governing power", according to a confidential document seen by Reuters.

"Basic democratic principles and checks and balances will be largely undermined, the legislature, law enforcement and media will be robbed of their independence and misused as a political arm, Big Tech will be given co-governing power,"

[#]uspol #politics #trump

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-16 at 19:40

David Lynch 1946-2025

"We are like the spider. We weave our life and then move along in it. We are like the dreamer who dreams and then lives in the dream. This is true for the entire universe."

– Brihadaranyaka Upanishad (sometimes quoted by David Lynch)

https://www.youtube.com/watch?v=bTGhEzBkqA0&list=OLAK5uy_mXoCszLQYRU4o5HquCcT2bLdWELuNIISY&index=11

[#]Lynch #DavidLynch #music

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-14 at 13:18

Plants the world over are absorbing about 31% more carbon dioxide than previously thought.

Cornell University & Oak Ridge National Laboratory used new models and measurements to assess GPP (Gross Primary Production) from the land at 157 petagrams of carbon per year, up from an estimate of 120 petagrams established 40 years ago and currently used in most estimates of Earth’s carbon cycle.

Terrestrial photosynthesis inferred from plant carbonyl sulfide uptake

https://www.nature.com/articles/s41586-024-08050-3

Press release: Plant CO2 uptake rises by nearly one third in new global estimates https://www.ornl.gov/news/plant-co2-uptake-rises-nearly-one-third-new-global-estimates

I have questions: If we increase CO2 but available nutrients stay the same, how it affects plants and herbivores?

Is the situation analogous to humans who eat too much carbs but don't get enough nutrients? Less nutrients per kg? Does vulnerability to plant diseases increase?

[#]climateChange #CO2 #botany #biology

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-13 at 02:10

Answer: Theoretical physics, Cosmohedra, a geometric object underlying the cosmological wavefunction for Tr(ϕ3) theory.

Cosmohedra

https://arxiv.org/abs/2412.19881

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-12 at 16:48

Sunday puzzle. Guess the field of study just by looking at illustrations.

[#]mathematics #geometry

=> View attached media | View attached media

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-12 at 14:25

A Movement That’s Quietly Reshaping Democracy For The Better

Citizens’ assemblies can help us better address societal challenges, overcome polarization and strengthen trust.

https://www.noemamag.com/a-movement-thats-quietly-reshaping-democracy-for-the-better/

[#]democracy #DeliberativeDemocracy

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-12 at 09:31

Who would have won the Simon-Ehrlich bet over different decades, and what do long-term prices tell us about resource scarcity?

https://ourworldindata.org/simon-ehrlich-bet

[#]economics #growth

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-11 at 12:00

Xinyu Guan et al, rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking, arXiv (2025).

https://arxiv.org/abs/2501.04519

Microsoft introduces rStar-Math, an SLM for math reasoning and problem solving

https://techxplore.com/news/2025-01-microsoft-rstar-math-slm-problem.html

"small language models (SLMs) can rival or even surpass the math reasoning capability of OpenAI o1, without distillation from superior models. rStar-Math achieves this by exercising "deep thinking" through Monte Carlo Tree Search (MCTS), where a math policy SLM performs test-time search guided by an SLM-based process reward model."

[#]AI #SLM #LLM

=> View attached media

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-07 at 09:55

NVIDIA Project DIGITS: A Grace Blackwell AI Supercomputer on your desk. https://www.nvidia.com/en-us/project-digits/

$3000 seems like a reasonable price for 1 petaflop FP4 home research workstation if you run it 24/7. Assuming electricity cost ~550 (€/$) per year, amortized over five years (3000+5×550) / (5×8760h) = 0.13 cents per hour.

[#]workstation #nvidia #linux #GPU #compute #deepLearning #AI

=> View attached media

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2025-01-04 at 10:39

Elektročas HH3 - the most accurate pendulum clock on the planet

https://dvaluch.web.cern.ch/hh3/

=> View attached media

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-26 at 19:55

Microsoft and OpenAI have contractual definition of AGI.

https://techcrunch.com/2024/12/26/microsoft-and-openai-have-a-financial-definition-of-agi-report/ The need for definition is because Microsoft loses access to OpenAI’s technology when the startup reaches AGI.

"The two companies reportedly signed an agreement last year stating OpenAI has only achieved AGI when it develops AI systems that can generate at least $100 billion in profits."

That's Saudi Aramco, Apple, Alphabet, level profits. https://en.wikipedia.org/wiki/List_of_largest_corporate_profits_and_losses

Currently OpenAI loses billions annually, and tells investors it won’t turn a profit until 2029.


I think Microsoft made a good bet.

[#]AGI #AI #Microsoft #OpenAI #business #economics

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-25 at 13:57

Comparing cooperative geometric puzzle solving in ants versus humans https://www.youtube.com/watch?v=ZHpu7ngQxwE

"Our results exemplify how simple minds can easily enjoy scalability while complex brains require extensive communication to cooperate efficiently."

https://www.pnas.org/doi/10.1073/pnas.2414274121

"we compared how individuals and groups of either ants or people tackle an identical geometrical puzzle. We find that when ants work in groups, their performances rise significantly. Groups of people do not show such improvement and, when their communication is restricted, even display deteriorated performances."

[#]cognition #swarmIntelligence #collectiveIntelligence #cooperation #biology #ants

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-23 at 20:50

Commercial tea bags release millions of microplastics when in use

https://www.uab.cat/web/newsroom/news-detail/-1345830290613.html?detid=1345940427095

polymer-based commercial tea bags release millions of nanoplastics and microplastics when infused. The study shows for the first time the capacity of these particles to be absorbed by human intestinal cells, and are thus able to reach the bloodstream and spread throughout the body.

Banaei G, Abass D, Tavakolpournegari A, Martín-Pérez J, Gutiérrez J, Peng G, Reemtsma T, Marc R, Hernández A, García-Rodríguez A. Teabag-derived micro/nanoplastics (true-to-life MNPLs) as a surrogate for real-life exposure escenaris. Chemosphere. 2024. 368:143736. doi: 10.1016/j.chemosphere.2024.143736.

https://www.sciencedirect.com/science/article/pii/S0045653524026377?via%3Dihub

Sorry for ruining your teatime. Loose tea users may hug and kiss the nearest tea caddy.

[#]tea #microplastics

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-23 at 10:54

REVIEW Twenty years of microplastic pollution research—what have we learned? https://www.science.org/doi/10.1126/science.adl2746 "The environmental burden of microplastics continues to grow, so a combination of scientific, economic, and social interventions will be necessary to curb that growth."

OECD REPORT Policy Scenarios for Eliminating Plastic Pollution by 2040 https://www.oecd.org/en/publications/policy-scenarios-for-eliminating-plastic-pollution-by-2040_76400890-en.html "Global growth in plastics production and use will continue to outpace population growth."

[#]microplastics #pollution #health #environment #OECD #policy

=> View attached media | View attached media

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-22 at 09:50

A synthetic protein-level neural network in mammalian cells

https://www.science.org/doi/10.1126/science.add8468

Bringing neural networks to life

A synthetic protein-based winner-take-all neural network controls cell fate decisions

https://www.science.org/doi/10.1126/science.adu1327

Chen et al. describe a protein-level synthetic circuit framework that implements a “winner-take-all” neural network capable of classifying the relative abundance of multiple inputs to modify the circuit output. This network replicates various classification circuits by adjusting the relative concentrations of a few components. By connecting this circuit to molecular pathways that regulate apoptosis, the authors demonstrate the promise of this approach for programming cell fate outcomes. This could be extended to design complex neural circuits that augment the computational capacity of cells.

perceptein: a combination of protein and perceptron.

biorxiv:

https://www.biorxiv.org/content/10.1101/2022.07.10.499405v1.full.pdf

https://phys.org/news/2024-12-perceptein-protein-based-artificial-neural.html

[#]neuralNetworks #biochem #systemsbiology #proteins #cellbiology

[#]bioengineering #technology

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-21 at 12:13

Shape, Symmetries, and Structure: The Changing Role of Mathematics in Machine Learning Research by Henry Kvinge

https://thegradient.pub/shape-symmetry-structure/

Maths and statistics struggle to provide immediate insight into the latest ML breakthroughs. The role of mathematics is evolving. It's use is shifting to higher-level such as matching architecture to underlying task structure or data symmetries..

In other worlds, fiber your bundles.

Some recent papers from the author seem relevant to the essay:

Machines and Mathematical Mutations: Using GNNs to Characterize Quiver Mutation Classes https://arxiv.org/abs/2411.07467

Machine Learning meets Algebraic Combinatorics: A Suite of Datasets to Accelerate AI for Mathematics Research

https://openreview.net/pdf?id=KQ1gI5qzAf

What Makes a Machine Learning Task a Good Candidate

for an Equivariant Network

https://openreview.net/pdf?id=46vfUIfIo1

[#]ML #MachineLearning #mathematics

[#]categoryTheory

=> View attached media

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-21 at 10:10

The Power of Norms

https://xphi.net/2024/12/12/the-power-of-norms/

A Common view about the power of norms is that they work by having an impact on beliefs. People conclude that certain options are bad.

Knobe argues that norms work by removing possibilities that can occur in the mind. When an option violates a norm, people tend not to think about it all.

"When we are truly in the grip of a norm, it’s not just that the norm impacts what we think of an option – it’s that it impacts which options we even think of at all."

[#]philosophy #psychology

=> More informations about this toot | View the thread

Written by ma𝕏pool on 2024-12-19 at 13:21

The Dimensions of dimensionality

https://www.cell.com/trends/cognitive-sciences/fulltext/S1364-6613(24)00189-X#

Globally interpretable feature dimensions are one property, among others like prediction performance and compactness, that researchers can prioritize when inferring latent representations.

Multidimensional representational spaces can capture complex structures such as hierarchical relationships, lexical entailment, and compositional features.

Superficially different representations, such as graphs and multidimensional data, can capture the same structural relationships under appropriate settings.

The dimensionality of a representation conveys relatively little information on its own.

[#]representation #dimensionality #cogsci #ml #datascience

=> View attached media | View attached media

=> More informations about this toot | View the thread

=> This profile with reblog | Go to maxpool@mathstodon.xyz account

Proxy Information
Original URL
gemini://mastogem.picasoft.net/profile/109314481661514525
Status Code
Success (20)
Meta
text/gemini
Capsule Response Time
524.633857 milliseconds
Gemini-to-HTML Time
14.149988 milliseconds

This content has been proxied by September (3851b).