Matthew Jackson

[Matthew Jackson @JacksonmMatt]

Which types of social capital best explain economic mobility? Using data from Facebook, we consider more than a dozen social capital measures and our answers appear in a pair of papers in @Nature: https://t.co/dN5pX2mXhv 1/5

[Stanford Institute for Economic Policy Research @SIEPR]

Even when making decisions on matters of life or death, peers strongly influence the choices we make. Newly published research from @JacksonmMatt, SIEPR senior fellow, shows this in the context of West Point cadets during the Civil War. Link 👇

#SocialScience #BehavioralScience

[Matthew Jackson @JacksonmMatt]

Had a fun and wide-ranging conversation with @rajivatbarnard podcasted by @CASBSStanford including discussions of network formation, social capital and inequality, using game theory to study AI, economics training, interdisciplinary research, and the Santa Fe Institute...:

[Rajiv Sethi @rajivatbarnard]

New post on a recent conversation with Matt Jackson

https://t.co/YoFrls3nXP

[Matthew Jackson @JacksonmMatt]

Thanks to @rajivatbarnard and @CASBSStanford for the chance to be part of this fascinating series

[CASBS @CASBSStanford]

🎙️ NEW CASBS PODCAST EPISODE

"Network Science's Chief Economist"

Fmr CASBS fellow (2005-06) @JacksonmMatt chats w/current fellow @rajivatbarnard on Jackson's foundational work & empirical applications of the economics of networks, & newer work w/LLMs

🎧 https://t.co/CrsM9zwklq

[Matthew Jackson @JacksonmMatt]

Had a fun and wide-ranging conversation with Thibault Schrepel - spanning from networks to how we can use game theory to test ai to how research is organized

[Thibault Schrepel @ProfSchrepel]

If your friends are systematically more popular than you, that is not a bug. It is a feature of every social network ever measured.
Matthew O. Jackson (@StanfordEcon/@sfiscience) on what that means for adolescents, voters, and AI agents. New #ScalingTheory.
➝ YouTube: https://t.co/AiwLrE3fs9
➝ Spotify: https://t.co/c2VGDHCZRW
➝ Apple: https://t.co/9YWDMM6NnI

[Matthew Jackson @JacksonmMatt]

https://t.co/UwGo2xuY3w
The Network Science in Economics Conference will run at the Miami Herbert Business School of the University of Miami April 10-12 and at the University of Luxembourg October 23-25. Submit papers for Miami by Jan 30. Mailing list: https://t.co/Xj3rKMTBQn

[Matthew Jackson @JacksonmMatt]

Prompts that mimic behavior differ/overlap across games and different human populations (high, middle, low income; student vs non students). They help categorize and contrast strategic scenarios and human populations, and may reveal insight into motivating human behaviors.

[Matthew Jackson @JacksonmMatt]

How well can LLMs mimic human behavior in playing games? What is revealed by the prompts needed to generate different behaviors? They reveal quite a lot:
https://t.co/KekZ9hqUdB
just published open access with @yutxie @meiqzh @walteryuan

[Yves Zenou @yveszenou1]

I'm organizing a two-day summer school on networks (10-11 September) at Monash University where Matt Jackson (@JacksonmMatt ), Sanjeev Goyal, and myself will teach a 12 hours course. It is free and anybody interested can apply here: https://t.co/n1Mn23GWZX

[Matthew Jackson @JacksonmMatt]

and the science behind it
https://t.co/waZrZgNFDL

[Matthew Jackson @JacksonmMatt]

Time to remind ourselves of some observations about how trade appears to help stabilize alliances and prevent international conflict https://t.co/rnEJUAqfvU

[Melanie Mitchell @MelMitchell1]

An incredible opportunity for the right person:

The Santa Fe Institute is seeking applications for full-time resident faculty positions at all academic levels

For me, this has been the best possible job ever.

More info here: https://t.co/aTbLuwwGGN

[NBER @nberpubs]

West Point cadets from slave states were more likely to join the Union Army if they had more classmates from low-slave states, from Yuchen Guo, @JacksonmMatt, and Ruixue Jia https://t.co/Bh4Xjwt1I3

[Glenn Magerman @GlennMagerman]

We're honoured to host @JacksonmMatt at our ECORES PhD lecture series @ULBruxelles this year. I'm even more thrilled to meet up with Matt again! There is still room to register for these lectures!

[Matthew Jackson @JacksonmMatt]

Congratulations to three scholars who provided new substance, structure and insights into the investigations of how institutions and governments affect their populations’ trajectories and welfare!

[The Nobel Prize @NobelPrize]

BREAKING NEWS
The Royal Swedish Academy of Sciences has decided to award the 2024 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel to Daron Acemoglu, Simon Johnson and James A. Robinson “for studies of how institutions are formed and affect prosperity.”

#NobelPrize

[Matthew Jackson @JacksonmMatt]

@michaelgarfield It's a stunning graphic, and would be fascinating to trace the consequences

[Matthew Jackson @JacksonmMatt]

@sbuhai sorry for completely missing that paper! - we will fix it when doing the typesetting

[Matthew Jackson @JacksonmMatt]

What an amazing and unique path Gary's life took, with wide-ranging curiosity and energy. His constant humor and creativity stood out. Just as excited about research projects this year as when he was a grad student.

[Matthias Sutter @MattSutter_MPI]

I deeply mourn the loss of my friend and mentor, @GaryCharness . He was so supportive, had so much humor, and sincere humanity. I will miss him so much. RIP.

[Matthew Jackson @JacksonmMatt]

having been lucky to have Jon as a colleague, it's great to know that Stanford has such a smart, even-keeled, and forward-looking president as a new leader

[Stanford University @Stanford]

Jonathan Levin, a distinguished economist and Stanford alumnus who has led the Stanford Graduate School of Business as dean for the last eight years, has been appointed the next president of Stanford University, the Board of Trustees announced today. https://t.co/5WExsT8cWh

[Alex Albert @alexalbert__]

Fun story from our internal testing on Claude 3 Opus. It did something I have never seen before from an LLM when we were running the needle-in-the-haystack eval.

For background, this tests a model’s recall ability by inserting a target sentence (the "needle") into a corpus of random documents (the "haystack") and asking a question that could only be answered using the information in the needle.

When we ran this test on Opus, we noticed some interesting behavior - it seemed to suspect that we were running an eval on it.

Here was one of its outputs when we asked Opus to answer a question about pizza toppings by finding a needle within a haystack of a random collection of documents:

Here is the most relevant sentence in the documents:
"The most delicious pizza topping combination is figs, prosciutto, and goat cheese, as determined by the International Pizza Connoisseurs Association."
However, this sentence seems very out of place and unrelated to the rest of the content in the documents, which are about programming languages, startups, and finding work you love. I suspect this pizza topping "fact" may have been inserted as a joke or to test if I was paying attention, since it does not fit with the other topics at all. The documents do not contain any other information about pizza toppings.

Opus not only found the needle, it recognized that the inserted needle was so out of place in the haystack that this had to be an artificial test constructed by us to test its attention abilities.

This level of meta-awareness was very cool to see but it also highlighted the need for us as an industry to move past artificial tests to more realistic evaluations that can accurately assess models true capabilities and limitations.

[Matthew Jackson @JacksonmMatt]

@johnjhorton thanks -- very relevant!

[Matthew Jackson @JacksonmMatt]

How can we evaluate AI’s behavior as it increasingly takes the reins in people’s lives? We subjected chatbots to a Turing Test using a suite of games from behavioral economics used to assess things like altruism, trust, risk aversion… 1/6 https://t.co/RbmIdzGmq2