Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile
Andreas Bjerre-Nielsen

@andbjn

associate prof. @Economics_UCPH @cph_sodas | topics: education & algorithms | methods: econ. theory, stats & machine learning | dad

ID: 307346772

linkhttp://bjerre-nielsen.me calendar_today29-05-2011 13:35:14

334 Tweet

710 Followers

415 Following

Tamar Haspel (@tamarhaspel) 's Twitter Profile Photo

Researchers pittted ChatGPT against genuine doctors, using actual patient questions. ChatGPT was more accurate and more empathetic. By a lot. HT Adam Cifu jamanetwork.com/journals/jamai…

Researchers pittted ChatGPT against genuine doctors, using actual patient questions.

ChatGPT was more accurate and more empathetic.

By a lot.  

HT <a href="/adamcifu/">Adam Cifu</a> 
jamanetwork.com/journals/jamai…
Kelton Minor (@keltonminor) 's Twitter Profile Photo

🚨Youth today spend an increasing fraction of their lives on mobile devices. Do natural settings🌳help urban youth disconnect from digital demands📲? Our new study logged young adult screen-time and green-time over two years to find out… 1/12 journals.sagepub.com/doi/10.1177/00…

Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile Photo

Last week I received a research grant to push forward research on school admission. We will study Danish high school admissions which provides an excellent setting for our research. Looking forward to get started on this project with Mikkel Gandil ! Stay tuned - soon we’re hiring!

Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile Photo

Market design for matching and school choice has focused on the manipulation of preferences and optimal assignment. Our new research shows that the admission criteria may affect whether students manipulate their admission eligibility which has wide consequences for school access.

Dean Eckles (@deaneckles) 's Twitter Profile Photo

People in some places in the US have more long ties — ties between people without mutual contacts. This is a map from our newly published paper: pnas.org/doi/10.1073/pn… Why does this matter? How do people end up with these different network structures? 🧵

People in some places in the US have more long ties — ties between people without mutual contacts.

This is a map from our newly published paper: pnas.org/doi/10.1073/pn…

Why does this matter? How do people end up with these different network structures? 🧵
Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile Photo

Does lower quality mean better research in economics? @peterbergman_ makes a laffer-curve analogy for improving publishing. He argues that lower acceptance thresholds make the knowledge disseminate faster and access to publishing more equitable.

Market Design Community (@econd47) 's Twitter Profile Photo

🚨Three weeks left to apply for this PhD position on applied school choice in Copenhagen 🇩🇰 with Andreas Bjerre-Nielsen and Mikkel Gandil🚨 All info: jobportal.ku.dk/phd/?show=1595… 👇👇👇 x.com/andbjn/status/…

Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile Photo

Johanna presented our exciting research project on how policies of social exposure within smaller and larger groups foster social interactions between students and its effect on homophily and emergence of network clustering (network triads). Stay tuned for more updates!

Pioneer Centre for AI (@aicentredk) 's Twitter Profile Photo

Open position at the Pioneer Centre for AI affiliated with DTU Compute as Associate Professor or DTU Tenure Track Assistant Professor in AI. Apply before 1 November, 2023. Read more about the position and apply here: efzu.fa.em2.oraclecloud.com/hcmUI/Candidat…

Open position at the Pioneer Centre for AI affiliated with <a href="/DTU_Compute/">DTU Compute</a> as Associate Professor or DTU Tenure Track Assistant Professor in AI.
Apply before 1 November, 2023.

Read more about the position and apply here: efzu.fa.em2.oraclecloud.com/hcmUI/Candidat…
Felix Chopra (@felixchopra) 's Twitter Profile Photo

1/ Qualitative interviews offer unparalleled richness but are rarely used in economics. Let's change that! New WP with Ingar Haaland uses an AI-driven approach to conducting q qualitative interviews, making them scalable, cheap, and ripe for both qualitative and quantitative analysis!

1/ Qualitative interviews offer unparalleled richness but are rarely used in economics. Let's change that! New WP with <a href="/Ingar30/">Ingar Haaland</a> uses an AI-driven approach to conducting q qualitative interviews, making them scalable, cheap, and ripe for both qualitative and quantitative analysis!
Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile Photo

This week at SODAS, Copenhagen (Bluesky: @cphsodas.bsky.social) you can hear about how techniques from eg ChatGPT can help to model and understand human behavior and careers. The talk is by Keyon Vafa in joint work with Susan Athey and David Blei

Prakash (Ate-a-Pi) (@8teapi) 's Twitter Profile Photo

Vicious Self-Degradation > you Google > Quora spots query and id’s as frequent > Quora uses ChatGPT to generate answer > ChatGPT hallucinates > Google picks up Quora answer as highest probability correct answer > ChatGPT hallucination is now canonical Google answer

Vicious Self-Degradation 

&gt; you Google 
&gt; Quora spots query and id’s as frequent
&gt; Quora uses ChatGPT to generate answer 
&gt; ChatGPT hallucinates 
&gt; Google picks up Quora answer as highest probability correct answer 
&gt; ChatGPT hallucination is now canonical Google answer
Tommy Andersson (@tommyan53041153) 's Twitter Profile Photo

Hey #econtwitter and job market candidates, we are recruiting at Department of Economics, Lund this year 👇I am chairing the committee for the position in "microeconomics (applied or theory) or financial Economics". Please apply!

Yohan (@yohaniddawela) 's Twitter Profile Photo

Meta’s Segment Anything Model (SAM) is the ChatGPT of geospatial data. It’s transforming geospatial analyses. Here’s what you need to know:

Andreas Bjerre-Nielsen (@andbjn) 's Twitter Profile Photo

Very happy that today I received a Starting Grant from DK Frie Forsk.fond. Together with an amazing group (Laust H. Mortensen, Anna Rogers, René Kizilcec), we will explore the potential for using new digital technology to measure learning and better understand learning.