Robert Hawkins

computational cognitive science researcher trying to understand how we understand each other | currently postdoc @princeton | starting 2023 at @UWMadison | #CogSci #Psychology #Communication #NLProc #Language

Robert Hawkins boosted:
Federico Pianzolafpianz@mstdn.social
2023-02-03

Postdoc position in graph reasoning for narrative
rug.nl/about-ug/work-with-us/j
Deadline: 19 Feb

Tasks: ontology learning, semantic relation identification, extracting structured information from narrative texts. #NLProc #DH #GNN #wikidata #fanfic #job #academic
@gronlp

Robert Hawkins boosted:
Aram Sinnreicharamsinn
2023-02-03

You don't have to be a Harvard professor to do the math on this one... 's pathbreaking work on and is a liability for a university that gets half a billion dollars from the owner of the greatest disinformation machine ever built. @commodon @communicationscholars @ICAHDQ @HigherEdLabor @academicchatter

Screenshot from Harvard Magazine: Chan Zuckerberg Commits $500 Million to Harvard Neuroscience and AI InstituteScreen shot of Washington Post headline: Harvard is shutting down project that studied social media misinformation. The school's Technology and Social Change Project, which published research on covid misinformation and the Jan. 6 Capitol riot, will end in 2024
Robert Hawkins boosted:
2023-01-21

Simpson's Paradox as a Simpsons paradox #xp

Robert Hawkins boosted:
2023-01-21

Do you want to do a psychology experiment while following best practices in open science? My collaborators and I have created Experimentology, a new open web textbook (to be published by MIT Press but free online forever).

experimentology.io

Some highlights! 🧵

Robert Hawkins boosted:
Sanjay Srivastavasanjaysrivastava@mas.to
2023-01-16

A fantastic way to make an important point. Researchers took items from widely-used scales of "social media addiction," and reworded them to be about spending time offline with friends. Using parallel criteria, they discovered that 69% of people are addicted to spending time with friends.

"Readers should approach our measure with a level of skepticism that should be afforded to current social media addiction measures."

(h/t @cconrymurray)

link.springer.com/article/10.3

Robert Hawkins boosted:
2023-01-15

Starting 2023, four universities are pausing or ending their Elsevier subscription due to exorbitant pricing.

Elsevier's subscription was costing them ~10% of their Libraries' entire budget.

"Elsevier’s prices have increased each year and have outpaced inflation"...this is despite Elsevier having the highest profit margins of virtually any other industry or publisher.

via dailyemerald.com/news/pressing
Figure via @MatteoCarandini
#OpenScience #AcademicPublishing #Science @academicchatter

Elsevier is more profitable than any other industry. E.g., Tobacco has a profit margin of ~17% vs. Elsevier's 37% profit margin.
Robert Hawkins boosted:
2023-01-13

šŸ“£ New! Beyond Single-Mindedness: A Figure-Ground Reversal for the Cognitive Sciences

:doi: doi.org/10.1111/cogs.13230 (html)
:OpenAccess: is.gd/beyond_singlemindedness_ (pdf)

Writing as a transdisciplinary collective from across the cognitive sciences and beyond, we highlight the opportunity for a Gestalt switch that puts interaction at the heart of cognition

Fedi authors include: @andreasliesenfeld @marlourasenberg @saulalbert @abebab @cdutilhnovaes @fusaroli @davidschlangen @LizStokoe

Title page of a Cognitive Science paper:
Beyond Single-Mindedness: A Figure-Ground Reversal for the Cognitive Sciences 

Mark Dingemanse, Andreas Liesenfeld, Marlou Rasenberg, Saul Albert, Felix K. Ameka, Abeba Birhane, Dimitris Bolis, Justine Cassell, Rebecca Clift, Elena Cuffari, Hanne De Jaegher, Catarina Dutilh Novaes, N.J. Enfield, Riccardo Fusaroli, Eleni Gregoromichelaki, Edwin Hutchins, Ivana Konvalinka, Damian Milton, Joanna Rączaszek-Leonardi, Vasudevi Reddy, Federico Rossano, David Schlangen, Johanna Seibt, Elizabeth Stokoe, Lucy Suchman, Cordula Vesper, Thalia Wheatley, Martina Wiltschko
Robert Hawkins boosted:
2023-01-13

"But let’s be clear: this ā€œburnoutā€ that secure scholars are feeling is phantom pain where their colleagues should be. Or, to use a term that every other normal worker in the US uses to describe their workplace under these conditions: you are suffering from the effects of intentional systemic understaffing." contingentmagazine.org/2023/01

Robert Hawkins boosted:
Roger Levyroger_p_levy
2023-01-13

Applications are open (due Feb 15) for MIT Brain & Cognitive Sciences' Post-Baccalaureate Research Scholars Program! This is a two-year, fully funded, program offering additional research and academic training in cognitive, systems or cellular neuroscience, cognitive science, computational cognitive science, computational neuroscience, or neuroengineering to individuals from under-represented groups and economically disadvantaged backgrounds. Please circulate widely! bcs.mit.edu/diversity-equity-a

Robert Hawkins boosted:
2023-01-12

My homage to Douglas Hofstadter's image, featured on the front cover of some editions of his book Gƶdel, Escher, Bach. He calls this a "triplet" sculpture, I believe because there are "trip" (three) "let" (letters).

Instead of the letters "G", "E", and "B", I used "3", "D", and "P", for "3D printing". #3DPrinting #Shadow #Art

A 3d printed sculpture hung from fishing wire, and lit by three spotlights, one along each of the x-, y-, and z-axes. The shadows cast show the symbols "3", "D", and "P".
Robert Hawkins boosted:
2023-01-12

ORBIT: A Unified Simulation Framework for Interactive Robot Learning Environments

abs: arxiv.org/abs/2301.04195
project page: isaac-orbit.github.io/

Robert Hawkins boosted:
Fahim Farookf@sigmoid.social
2023-01-11

"CREPE: Can Vision-Language Foundation Models Reason Compositionally?. (arXiv:2212.07796v2 [cs.CL] UPDATED)" — A compositionality evaluation benchmark to measures two important aspects of compositionality — systematicity and productivity — across 6 architectures trained with 4 algorithms on massive datasets.

Paper: arxiv.org/abs/2212.07796

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too šŸ™‚>>

We introduce CREPE, a benchmark to evaluate whether vision-language foundation models demonstrate two fundamental aspects of compositionality: systematicity and productivity. To evaluate systematicity, CREPE utilizes Visual Genome and introduces three new test datasets for the three popular pretraining datasets: CC-12M, YFCC-15M, and LAION-400M. These enable evaluating models’ abilities to systematically generalize their understanding to seen compounds, unseen compounds, and even unseen atoms. To evaluate productivity, CREPE introduces examples of nine complexities, with three types of hard negatives for each.
Robert Hawkins boosted:
Bertrand CharpentierBertrand_Charp@sigmoid.social
2023-01-10

Yesterday, Yudong Luo presented his NeurIPS paper "Uncertainty-Aware RL for Risk-Sensitive Player Evaluation in Sports Game" at Uncertainty in AI reading group !

uncertainty-reading-group.gith

Co-author: Guiliang Liu, Oliver Schulte, Pascal Poupart

youtube.com/watch?v=QCAVcKnxbV

Robert Hawkins boosted:
2023-01-10

My first #neuroscience masto-print? Cross-posted to twitter, so apologies for the bird links throughout

Recently we proposed that ventral cortex (VTC) may not represent global shape, but a basis set of local features (twitter.com/vayzenberg90/statu)

In her commentary doi.org/10.1016/j.tics.2022.11, Yaoda Xu raises several important questions which we had the opportunity to address doi.org/10.1016/j.tics.2022.12

1/n

Robert Hawkins boosted:
2023-01-10

Hey #DH #DigitalHumanities #DigitalHistory #CulturalAnalytics #ComputationalSocialScience ppl! I'm teaching a course this Spring on Culture At Scale and would love to hear abt your favorite books or articles that make ✨knowledge claims✨ about culture with computational methods šŸ¤–šŸ™ !!

Could be from or about any time period, disciplinary/domain area, or in any programming language/methodological approach -- so please lmk what you really enjoyed! Thanks 😊 !!!

Robert Hawkins boosted:
2023-01-10

Great review out in @eLife from
@ElDuvelle @rmgrieves & Matt van der Meer:
Temporal context and latent state inference in the hippocampal splitter signal

elifesciences.org/articles/823

2023-01-10

@BayesForDays fr need to install a simulator in my actual bedroom

2023-01-10

i don’t think i’m an especially terrible participant but every time i help with an fmri study i come away convinced the only thing they could possibly be measuring is me exerting every ounce of willpower to keep my eyes open

Robert Hawkins boosted:
2023-01-04

Oh oh I cannot wait for The Little Learner to come out! mitpress.mit.edu/9780262546379

Robert Hawkins boosted:
2023-01-04

What makes a tutoring session more effective than a lecture? In our new preprint "Learning to teach, teaching to learn," we propose a hierarchical Bayesian model of pedagogy to explore the cognitive mechanisms underlying adaptive teaching.🧵

w/ Andrew Palacci @natvelali @rdhawkins @gershbrain

psyarxiv.com/4u5g6/

Client Info

Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst