Tiago Peixoto
@tiago.skewed.de
Statistical mechanic, secular Bayesian.
Prof. of Complex Systems and Network Science @ IT:U, Austria
Head of the “Inverse Complexity Lab”.
@invcomplexity.skewed.de
https://skewed.de/lab
Prof. of Complex Systems and Network Science @ IT:U, Austria
Head of the “Inverse Complexity Lab”.
@invcomplexity.skewed.de
https://skewed.de/lab
Pinned
Tiago Peixoto
@tiago.skewed.de
· Nov 5
Double feature, just out on @royalsocietypublishing.org:
"Uncertainty quantification and posterior sampling for network reconstruction"
Previous explainer thread here: bsky.app/profile/tiag...
Code here: graph-tool.skewed.de/static/docs/...
royalsocietypublishing.org/doi/10.1098/...
"Uncertainty quantification and posterior sampling for network reconstruction"
Previous explainer thread here: bsky.app/profile/tiag...
Code here: graph-tool.skewed.de/static/docs/...
royalsocietypublishing.org/doi/10.1098/...
Can't get over the fact that in the ML literature the act of *sampling* from an inferred model is now called “inference.”
We really are in the worst timeline.
We really are in the worst timeline.
December 8, 2025 at 8:00 PM
Can't get over the fact that in the ML literature the act of *sampling* from an inferred model is now called “inference.”
We really are in the worst timeline.
We really are in the worst timeline.
Reposted by Tiago Peixoto
I wrote a blog post about the often stated but never explained assumption that communities in graphs should always be connected.
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
The perplexing “connected cluster axiom” – Inverse Complexity Lab
Research group on inverse problems in complex systems and network science.
skewed.de
December 4, 2025 at 9:10 PM
I wrote a blog post about the often stated but never explained assumption that communities in graphs should always be connected.
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
Reposted by Tiago Peixoto
University is only fee free in a small handful of countries. Source: buff.ly/PAuDqrF
December 7, 2025 at 4:30 PM
University is only fee free in a small handful of countries. Source: buff.ly/PAuDqrF
Reposted by Tiago Peixoto
December 7, 2025 at 7:09 AM
I wrote a blog post about the often stated but never explained assumption that communities in graphs should always be connected.
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
The perplexing “connected cluster axiom” – Inverse Complexity Lab
Research group on inverse problems in complex systems and network science.
skewed.de
December 4, 2025 at 9:10 PM
I wrote a blog post about the often stated but never explained assumption that communities in graphs should always be connected.
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
This is inconsistent with statistical significance and null models that underlie the most widely employed methods.
skewed.de/lab/posts/co...
Reposted by Tiago Peixoto
All super rich guys sound the same now. It’s all “I sleep four hours a night and spend the other twenty hours a day developing a product that’s going to bring us the best widespread poverty anyone’s ever seen.”
December 3, 2025 at 9:53 PM
All super rich guys sound the same now. It’s all “I sleep four hours a night and spend the other twenty hours a day developing a product that’s going to bring us the best widespread poverty anyone’s ever seen.”
Just a thought: how about substantial investment and subsidy for mass transportation instead?
I agree: there’s a public health imperative to quickly expand the adoption of autonomous vehicles, which will save many lives. More than 39,000 Americans died in motor vehicle crashes yearly. www.nytimes.com/2025/12/02/o...
December 3, 2025 at 6:00 AM
Just a thought: how about substantial investment and subsidy for mass transportation instead?
Reposted by Tiago Peixoto
I don’t know if anyone else notices or cares, but when I see a presentation in which the speaker uses obviously generated-AI images to illustrate their slides, it makes me immediately less confident in whatever other content they’re presenting.
November 28, 2025 at 3:07 PM
I don’t know if anyone else notices or cares, but when I see a presentation in which the speaker uses obviously generated-AI images to illustrate their slides, it makes me immediately less confident in whatever other content they’re presenting.
Reposted by Tiago Peixoto
The UN General Assembly resolution against torture was rejected by only three countries: the US, Israel, and Argentina, who remind the world that they oppose even the most basic principles of humanity.
November 27, 2025 at 5:16 AM
The UN General Assembly resolution against torture was rejected by only three countries: the US, Israel, and Argentina, who remind the world that they oppose even the most basic principles of humanity.
“Then you tell the model, 'Can you please fix the bug?' [...] And it introduces a second bug. Then you tell it, “You have this new second bug,” [...] and brings back the first bug, and you can alternate between those. How is that possible?”
Because it's a fucking Markov chain on word tokens.
Because it's a fucking Markov chain on word tokens.
November 25, 2025 at 8:57 PM
“Then you tell the model, 'Can you please fix the bug?' [...] And it introduces a second bug. Then you tell it, “You have this new second bug,” [...] and brings back the first bug, and you can alternate between those. How is that possible?”
Because it's a fucking Markov chain on word tokens.
Because it's a fucking Markov chain on word tokens.
So, the ”age” of scaling was about 3 years or so?
November 25, 2025 at 8:51 PM
So, the ”age” of scaling was about 3 years or so?
Reposted by Tiago Peixoto
“Are we the baddies?”
November 24, 2025 at 3:24 AM
“Are we the baddies?”
Reposted by Tiago Peixoto
Getting journal rejections like
November 21, 2025 at 5:50 PM
Getting journal rejections like
Reposted by Tiago Peixoto
I don’t want to seem out of touch but I don’t actually understand the economy anymore.
November 18, 2025 at 3:23 AM
I don’t want to seem out of touch but I don’t actually understand the economy anymore.
Reposted by Tiago Peixoto
And so the genocide continues in hushed silence, with the rest of the world lulled into a false sense that it has ended... www.theguardian.com/world/2025/n...
US military planning for divided Gaza with ‘green zone’ secured by international and Israeli troops
Exclusive: Almost all Palestinians have been displaced to ‘red zone’ where no reconstruction is planned
www.theguardian.com
November 14, 2025 at 9:30 PM
And so the genocide continues in hushed silence, with the rest of the world lulled into a false sense that it has ended... www.theguardian.com/world/2025/n...
This level of blatant dishonesty baffles me. I'm seeing this more and more. Do people *want* to ruin their reputation?
Doing non-causal inference (and being explicit about it), yet using a causal word as second word in the title.
If you pay Nature € 10.690, they will publish this in Nature Ageing.
I can tell you what I think of that for free.
www.nature.com/articles/s43...
If you pay Nature € 10.690, they will publish this in Nature Ageing.
I can tell you what I think of that for free.
www.nature.com/articles/s43...
November 14, 2025 at 7:21 AM
This level of blatant dishonesty baffles me. I'm seeing this more and more. Do people *want* to ruin their reputation?
Reposted by Tiago Peixoto
You know what they say about academics...
"causal" in the streets
"preliminary and hypothesis-generating" in the sheets
"causal" in the streets
"preliminary and hypothesis-generating" in the sheets
November 12, 2025 at 11:16 PM
You know what they say about academics...
"causal" in the streets
"preliminary and hypothesis-generating" in the sheets
"causal" in the streets
"preliminary and hypothesis-generating" in the sheets
“Metros reduce car use in European cities but trams do not”
The conclusion says: “Although cities with a metro tend to
have a higher share of B, it is unclear whether the presence itself causes people to shift to B, and no causal claims can be made.”
Then, why make a causal claim in the title? 🤷♂️
The conclusion says: “Although cities with a metro tend to
have a higher share of B, it is unclear whether the presence itself causes people to shift to B, and no causal claims can be made.”
Then, why make a causal claim in the title? 🤷♂️
I analysed the modal share of nearly 400 European cities, comparing those with a metro, only a tram, and without any rail system.
Large cities with a metro have roughly half the car share of cities served only by a tram.
www.nature.com/articles/s44...
Data: CitiesMoving.com
Large cities with a metro have roughly half the car share of cities served only by a tram.
www.nature.com/articles/s44...
Data: CitiesMoving.com
November 11, 2025 at 9:48 AM
“Metros reduce car use in European cities but trams do not”
The conclusion says: “Although cities with a metro tend to
have a higher share of B, it is unclear whether the presence itself causes people to shift to B, and no causal claims can be made.”
Then, why make a causal claim in the title? 🤷♂️
The conclusion says: “Although cities with a metro tend to
have a higher share of B, it is unclear whether the presence itself causes people to shift to B, and no causal claims can be made.”
Then, why make a causal claim in the title? 🤷♂️
Reposted by Tiago Peixoto
A statistic that concentrates the mind: In the first 6 months of 2025, China installed 250 GW of solar power capacity. This is around 150% of France's aggregate electricity production...
November 11, 2025 at 8:27 AM
A statistic that concentrates the mind: In the first 6 months of 2025, China installed 250 GW of solar power capacity. This is around 150% of France's aggregate electricity production...
Firing a BBC director for *leftist* bias is like punishing a pig for taking a bath.
Tim Davie resigns as BBC director general after accusations of ‘serious and systemic’ bias in coverage
Tim Davie resigns as BBC director general after accusations of ‘serious and systemic’ bias in coverage
Davie tells staff departure was “my own decision” but it comes with the BBC preparing to apologise for the way it edited a Trump speech
Tim Davie, the BBC’s director general, and the head of BBC News have resigned after a former adviser to the corporation accused it of “serious and systemic” bias in its coverage of issues including Donald Trump, Gaza and trans rights.
In a shock move, Davie said his depature was his own, but comes with the BBC preparing to apologise for the way the broadcaster edited a Donald Trump speech. Continue reading...
www.theguardian.com
November 9, 2025 at 7:22 PM
Firing a BBC director for *leftist* bias is like punishing a pig for taking a bath.
A team of 42 researchers lead by Oxford finds that bears do, in fact, defecate in the woods.
www.oii.ox.ac.uk/news-events/...
www.oii.ox.ac.uk/news-events/...
OII | Study identifies weaknesses in how AI systems are evaluated
Largest systematic review of AI benchmarks highlights need for clearer definitions and stronger scientific standards.
www.oii.ox.ac.uk
November 9, 2025 at 9:33 AM
A team of 42 researchers lead by Oxford finds that bears do, in fact, defecate in the woods.
www.oii.ox.ac.uk/news-events/...
www.oii.ox.ac.uk/news-events/...
James Watson is perhaps the greatest cautionary tale of where the competitive narcisism that poisons science can lead.
www.statnews.com/2025/11/07/j...
www.statnews.com/2025/11/07/j...
James Watson, dead at 97, was a scientific legend and a pariah among his peers
James Watson, the co-discoverer of the structure of DNA who died Thursday at 97, was a scientific legend and a pariah among his peers.
www.statnews.com
November 9, 2025 at 8:08 AM
James Watson is perhaps the greatest cautionary tale of where the competitive narcisism that poisons science can lead.
www.statnews.com/2025/11/07/j...
www.statnews.com/2025/11/07/j...
Very good post. I very much agree with the overall position.
How can we reform science? I have some ideas. But I am not sure you’ll like them, because they don’t promise much. elevanth.org/blog/2025/07...
Which Kind of Science Reform
What hope is there for science reform, if we can't agree on what to reform? Right now, principles are more important than practices.
elevanth.org
November 8, 2025 at 11:38 AM
Very good post. I very much agree with the overall position.
Nowadays, often the justification for choosing M$ is that it claims to be GDPR compliant, which just means that M$ can afford getting sued for any breach, which of course happens all the time. And universities don't care about potential breaches themselves, only about liability, which M$ absorbs.
We had this discussion 25 years ago.
A colleague calculated that they could hire a full-time administrator for Linux instead of paying the MS licences. Most thought of it as absurd idea ;-)
A colleague calculated that they could hire a full-time administrator for Linux instead of paying the MS licences. Most thought of it as absurd idea ;-)
November 6, 2025 at 4:11 PM
Nowadays, often the justification for choosing M$ is that it claims to be GDPR compliant, which just means that M$ can afford getting sued for any breach, which of course happens all the time. And universities don't care about potential breaches themselves, only about liability, which M$ absorbs.
Reposted by Tiago Peixoto
lol
November 6, 2025 at 4:47 AM
lol