Lightnews — Scholar-powered news

Eric Colson @ecolson.bsky.social · Aug 9

Really can't seem to stop watching glass fruit being cut!

Reposted by Eric Colson

Carl Quintanilla @carlquintanilla.bsky.social · Jun 22

A reminder of how the number of centrifuges in Iran soared after Trump’s 2018 JCPOA withdrawal.

@brendannyhan.bsky.social

38 430 1.1K

Eric Colson @ecolson.bsky.social · May 6

Wow - CA and NY pay out a lot more than they get back to the Feds.
usafacts.org/articles/whi...

Eric Colson @ecolson.bsky.social · Apr 18

Compelling visualization showing global temperature over time. It's the last bit that is compelling. Avg is about the same, but variation is increased.
fb.watch/z1Axq0JLi-/?...

1

Eric Colson @ecolson.bsky.social · Apr 17

Tariffs = Tax = Higher Prices.
I'm getting emails from ecommerce companies warning of price increases.

“It is the maxim of every prudent master of a family, never to attempt to make at home what it will cost him more to make than to buy.”
— Adam Smith, The Wealth of Nations
#tariffs

Eric Colson @ecolson.bsky.social · Mar 25

Interesting. How did you obtain this data?

1

Eric Colson @ecolson.bsky.social · Mar 21

Gaslight me on this:
www.economist.com/graphic-deta...

2

Eric Colson @ecolson.bsky.social · Mar 15

Fundamental. There are always options for decomposability in just about any system.

Rick Marriner @rickmarriner.com · Mar 15

Herbert Simon’s scientific “parable of the two watchmakers” illustrates how modular design enhances efficiency and resilience in complex systems.

Is this principle still relevant in today’s data science practices? It feels resilient and evolutionary.

#ComplexSystems #ModularDesign

II. Herbert A. Simon and "Near-Decomposability" — MENIMAGERIE

“Why is reality structured at all?” was what Herbert Simon, an extraordinary polymath regarded as one of last century’s most influential thinkers, asked himself in his watchmaker parable from the arti...

www.menimagerie.com

1

Eric Colson @ecolson.bsky.social · Mar 14

I'm also disappointed that people still use X. Tbh, bsky is not quite there yet in terms of network, content and engagement. But, it just takes a few more people to come over till it tips.

1

Eric Colson @ecolson.bsky.social · Mar 2

helpful. thanks for sharing.

1

Eric Colson @ecolson.bsky.social · Feb 26

Progress on the elusive sequence-based recommendations.
arxiv.org/pdf/2502.13909
#machinelearning, #MLSky, #DataScience

2

Eric Colson @ecolson.bsky.social · Feb 20

t-SNE?

1

Eric Colson @ecolson.bsky.social · Feb 18

What is Solopreneurship?

1 1

Eric Colson @ecolson.bsky.social · Feb 17

A shame. he creates a narrative of fraud that is unsubstantiated but will never be held accountable.

Reposted by Eric Colson

Towards Data Science @towardsdatascience.com · Feb 14

Quantum computing and data science... a match made in tech heaven? Sara A. Metwalli's latest article explores how these 2 fields intersect and why data scientists should care about quantum advancements.

Should Data Scientists Care About Quantum Computing? | Towards Data Science

I am sure the quantum hype has reached every person in tech (and outside it, most probably). With some over-the-top claims, like “some company has proved quantum supremacy,” “the quantum revolution…

towardsdatascience.com

1 1 7

Eric Colson @ecolson.bsky.social · Feb 14

Great use of AI (hope it's sustainable)
#ai

Greta Warren @gretawarren.bsky.social · Feb 14

How can explainable AI empower fact-checkers to tackle misinformation? 📰🤖

We interviewed fact-checkers & identified explanation needs & design implications for #NLP, #HCI, and #XAI.

Excited to present this work with
@iaugenstein.bsky.social & Irina Shklovski at #chi2025!
arxiv.org/abs/2502.09083

Eric Colson @ecolson.bsky.social · Feb 12

don't leave us in suspense. Do you have an explanation?

1 1

Eric Colson @ecolson.bsky.social · Feb 6

Operationalizing Machine Learning: An Interview Study by @joehellerstein.bsky.social, @adityagp.bsky.social, et al. Particularly love the part on "Retrofitting Explanations".
#MachineLearning #MLOps #Datascience.
arxiv.org/pdf/2209.09125

1 4 13

Eric Colson @ecolson.bsky.social · Feb 4

The good part of AI.

Scott McGrath @smcgrath.phd · Feb 4

🧪 🩺🖥️ #MLSky

Eric Topol @erictopol.bsky.social · Feb 4

New: The largest medical A.I. randomized controlled trial yet performed, enrolling >100,000 women undergoing mammography screening
The use of AI led to 29% higher detection of cancer, no increase of false positives, and reduced workload compared with radiologists w/o AI thelancet.com/journals/lan...

Reposted by Eric Colson

Scott McGrath @smcgrath.phd · Feb 4

🧪 🩺🖥️ #MLSky

Eric Topol @erictopol.bsky.social · Feb 4

New: The largest medical A.I. randomized controlled trial yet performed, enrolling >100,000 women undergoing mammography screening
The use of AI led to 29% higher detection of cancer, no increase of false positives, and reduced workload compared with radiologists w/o AI thelancet.com/journals/lan...

Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI): a randomised, controlled, parallel-group, non-inferiority, ...

The findings suggest that AI contributes to the early detection of clinically relevant breast cancer and reduces screen-reading workload without increasing false positives.

thelancet.com

4 16 76

Eric Colson @ecolson.bsky.social · Feb 4

Thoughtful article on the role of statistics in AI by Friedrich, et al. I didn't think there would be any question about it. Causal reasoning, uncertainty, evaluation, etc - all are crucial to AI and come from the school of statistics.
link.springer.com/article/10.1...
#DataScience #Statistics #AI

Is there a role for statistics in artificial intelligence? - Advances in Data Analysis and Classification

The research on and application of artificial intelligence (AI) has triggered a comprehensive scientific, economic, social and political discussion. Here we argue that statistics, as an interdisciplin...

link.springer.com

3 4

Eric Colson @ecolson.bsky.social · Feb 2

You pull 2 eggs. The chances your 1st pull is double yolked is 1/1,000. The chances for your 2nd pull is also 1/1,000. The chances BOTH pulls are double yolked is 1/1,000,000, since both have to happen (1/1,000 * 1/1,000). Beyond 2: (1/1000)^n, where n is the number of double yolked eggs to find.

2

Eric Colson @ecolson.bsky.social · Feb 1

Always an interesting conversation with @hugobowne.bsky.social. Great topics on the management of data scientists. #Datascience
www.linkedin.com/posts/hugo-b...

Hugo Bowne-Anderson on LinkedIn: 🎙️ Why 90% of Data Science Projects Fail—And How to Fix It In the…

🎙️ Why 90% of Data Science Projects Fail—And How to Fix It In the latest High Signal podcast, I spoke with Eric Colson—DS/ML advisor, former Chief…

www.linkedin.com

1 1 2

Eric Colson @ecolson.bsky.social · Jan 31

The Secretary Problem. An oldie, but so relevant to many business problems (and personal ... eg. when to stop looking for a parking spot).
www.jstor.org/stable/2243190
#Algorithms #DataScience

1st page of the paper with introduction.

1

Eric Colson @ecolson.bsky.social · Jan 21

This resonates ... for 20+ years, I've embraced ANN only as as a very lose metaphor for how our human brains work. But in the last 2 years I couldn't help but notice more and more similarities.

Scott McGrath @smcgrath.phd · Jan 21

🧪 Researchers from MIT, Harvard, and NYU found human brains and AI neural networks represent information similarly, despite distinct architectures. This suggests intelligence converges when solving shared challenges, like language or vision. 🩺💻 #MLSky

Universality of representation in biological and artificial neural networks

Many artificial neural networks (ANNs) trained with ecologically plausible objectives on naturalistic data align with behavior and neural representations in biological systems. Here, we show that…

www.biorxiv.org

1