Sign in

cogscikid.bsky.social

@cogscikid.bsky.social
301 followers 85 following 103 posts

Kempner Institute research fellow @Harvard I want to build AI assistants for the human reinforcement learning algorithm. prev: deepmind, umich, msr cogscikid.com

PostsRepliesMedia
Reposted by @cogscikid.bsky.social
Franziska Brändle @frabraendle.bsky.social · 29/09/2026
Incredibly excited to announce that I’m starting my own research group at Justus-Liebig-Universität in Gießen as part of the excellence cluster “The Adaptive Mind”! 🥳 I'm looking to fill the first PhD position. If you’re interested, apply here: tinyurl.com/38p3h22p Please spread the word! 🙏
jobs.uni-giessen.de
Doctoral student (m/f/d) Computational Motivation – Justus-Liebig-Universität Gießen
44522
Reposted by @cogscikid.bsky.social
RLDM @rldmparis2027.bsky.social · 29/09/2026
📣 Save the date! RLDM 2027, the Multi-disciplinary Conference on Reinforcement Learning and Decision Making, is coming to Paris, July 6-9, 2027 🎉 Stay tuned for submission deadlines and more updates on our website rldm.org 👀
06937
Reposted by @cogscikid.bsky.social
Cory Inman @brainscout.bsky.social · 05/09/2026
The Psychology Department at the University of Utah is currently seeking applicants for a Tenure-Track Assistant Professor position in Human Neuroscience. Please share the advertisement below with anyone in your professional network who is on the market! utah.peopleadmin.com/postings/206...
13321
Reposted by @cogscikid.bsky.social
Marc Coutanche @marccoutanche.bsky.social · 01/09/2026
We are recruiting an Assistant Professor in Cog Neuro/Cog Psych to join us at Rice! Please spread the word! apply.interfolio.com/192080
apply.interfolio.com
Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio
24642
Reposted by @cogscikid.bsky.social
Markus Brauer @markusbrauer.bsky.social · 30/08/2026
The UW–Madison Department of Psychology is seeking to fill an open-rank faculty position in social psychology. If you know a great candidate, encourage them to apply! jobs.wisc.edu/jobs/assista...
jobs.wisc.edu
Assistant Professor, Associate Professor, or Professor of Psychology - Madison, Wisconsin, United States
Current Employees: If you are currently employed at any of the Universities of Wisconsin, log in to Workday to apply through the internal application process.Job Category:FacultyEmployment Type:Regula...
02112
Reposted by @cogscikid.bsky.social
Continual RL Workshop @continual-rl.bsky.social · 26/08/2026
Our second contributed talk in our workshop was given by @cogscikid.bsky.social and Neil Shah: 🎙️Task diversity produces systematic transfer but inhibits continual reinforcement learning 📷 openreview.net/forum?id=3Wy...
051
Reposted by @cogscikid.bsky.social
Tim Kietzmann @timkietzmann.bsky.social · 19/08/2026
📢 We are hiring a full professor in "Intelligence in Biological and Artificial Systems". Come join us at Germany's first and largest cognitive science program. Please share far and wide, and get in touch with any questions. www.uni-osnabrueck.de/en/universit...
uni-osnabrueck.de
129 FB 8 W3 Professorship Intelligence in Biological and Artificial Systems, Institute of Cognitive Science: Uni Osnabrück
05731
Reposted by @cogscikid.bsky.social
Sam Nastase @samnastase.bsky.social · 21/08/2026
We’re recruiting a full-time lab manager to join the Shared Minds Lab at USC! This will be a great opportunity for someone who wants to get hands-on experience with research before starting a PhD program in psychology or neuroscience. More below:
44123
Reposted by @cogscikid.bsky.social
Megan deBettencourt @mdebettencourt.bsky.social · 17/08/2026
Come work with me at Google! We have an open 6-month temp role on my team - it’s a great opportunity for someone with a cog neuro or psych background interested in exploring big tech. Looking for comp modeling experience (e.g. RL, DDMs) and Python/JS. Reach out to learn more or pass along any names!
04230
Reposted by @cogscikid.bsky.social
Jamal A. Williams @jayneuro.bsky.social · 19/08/2026
Our new paper is out in Nature Communications! 🎶🧠✨ If you're interested in how the brain represents meaning evoked by music, check it out. Shout out to my Princeton team for making this happen!🚀 @jalon.bsky.social, @margulisa.bsky.social, Karen Christianson, Cara Turnbull, Grace Simmons, Uri Hasson
0166
Reposted by @cogscikid.bsky.social
Sam Gershman @gershbrain.bsky.social · 18/08/2026
@arthurpr4t.bsky.social has a new preprint with important results on a famous psychophysical law (Weber's law). It isn't, in fact, a law, because it can be broken. A more fundamental principle (efficient coding) shows when and why Weber's law holds true. www.biorxiv.org/content/10.6...
biorxiv.org
Efficient coding makes and breaks Weber's law
Weber's law is a rare quantitative regularity in psychology, yet its origins remain debated. Here we provide causal evidence that it arises from the more fundamental principle of efficient coding. Thi...
413041
Reposted by @cogscikid.bsky.social
404 Media @404media.co · 17/08/2026
EXCLUSIVE: 404 Media placed a tracking device in a shipment of rare books to see which AI company was buying it, and found an Amazon facility where Amazon scans and destroys books. @evystadium.bsky.social has more. Investigation by @emanuelmaiberg.bsky.social here: www.404media.co/we-tracked-a...
5727821762
Reposted by @cogscikid.bsky.social
Alireza Modirshanechi @modirshanechi.bsky.social · 15/08/2026
Excited to share that I'll be starting as a junior professor and Emmy Noether awardee at the University of Göttingen this October! 🥳 🚀 I'll be hiring #phd students and #postdoc across #machinelearning, #neuroscience and #cogsci: modirlab.github.io/open-positio... Please spread the word! 🙏
PhD ad informationPostdoc ad informationInformation about Göttingen and equal opportunities
1217059
Reposted by @cogscikid.bsky.social
Black In Neuro @blackinneuro.bsky.social · 12/08/2026
Applications are open for the 2026–27 Black In Neuro Mentorship Cohort! 🧠✨ Apply as a mentor or mentee for 1:1 matching, application support, professional development & a global neuro community. 📲 Scan the QR code to apply! #BlackInNeuro
Flyer for the Black In Neuro Mentorship Program announcing applications for the 2026–27 cohort. Participants can apply as mentors or mentees. Program benefits include 1:1 mentor matching, graduate/medical school application support, community, professional development workshops, and an honorarium. Two QR codes are provided: one for the mentor application and one for the mentee application. The flyer has a dark brown background with cream and gold accents and the Black In Neuro logo in the upper left. For questions, contact mentorship@blackinneuro.com
01312
Reposted by @cogscikid.bsky.social
Jack Mirkinson @jackmirkinson.bsky.social · 11/08/2026
Today in @thenation.com: a harrowing piece written from inside the Dilley ICE detention center by Maryam Tahmasebi, a green card holder who's been locked up along with her husband and son for months despite none of them committing any crime. www.thenation.com/article/soci...
thenation.com
One Woman’s Message From the Hell of ICE Detention
My husband, son, and I have been locked up for months—despite being permanent residents with no criminal record. It is an unendurable, unending nightmare.
11835222210
Reposted by @cogscikid.bsky.social
Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 12/08/2026
My politics at this point is just wanting justice for all the immigrants who have been so cruelly wronged
11267213
cogscikid.bsky.social @cogscikid.bsky.social · 11/08/2026
huggingface makes 1-click 1-line-of-code downloads and uploads really easy! and supports versioning!
010
cogscikid.bsky.social @cogscikid.bsky.social · 10/08/2026
naturalistic cog comp neuro lab! sounds cool!
020
Reposted by @cogscikid.bsky.social
Taylor Webb @taylorwwebb.bsky.social · 26/07/2026
I'm recruiting two postdocs (or potentially PhD students) to work on projects at the intersection of cognition, neuroscience, and AI (with a particular emphasis on mechanistic interpretability). Apply here: docs.google.com/forms/d/e/1F... and please share!
docs.google.com
37142
cogscikid.bsky.social @cogscikid.bsky.social · 28/06/2026
I think AI has tremendous potential to help us *help each other* This year I built an AI copilot to help support and scale mentorship. I plan to write a couple of blog posts on what I learned from this. Here's a link to the first one! infinitecare.substack.com/p/what-i-lea...
0152
Reposted by @cogscikid.bsky.social
Natalia Vélez @natvelali.bsky.social · 24/06/2026
It feels strange to share good news, but: I’m moving to UCSD in July!! 🥳 I'm very grateful for my time at Princeton and SO excited for what comes next. We're building multiplayer games to study how teams are shaped by the problems they solve. If that sounds fun to you, I'm accepting grad students!
Paragliders hover between the cliffs and sea in Torrey Pines Gliderport, La Jolla, CA, USA. Photo credit Micah Fluhart for Unsplash.
312116
cogscikid.bsky.social @cogscikid.bsky.social · 20/06/2026
📄 Paper: arxiv.org/abs/2606.00880 📷 Project page: cogscikid.com/banyan/ Feedback welcome!!
arxiv.org
Task diversity produces systematic transfer but inhibits continual reinforcement learning
Continual reinforcement learning aims to produce agents that learn not only to improve at their current tasks but also to adapt as task distributions change. Training an agent on many diverse tasks ca...
320
cogscikid.bsky.social @cogscikid.bsky.social · 20/06/2026
Excited to share this work has been selected for an Oral at the Reinforcement Learning Conference Continual RL workshop! @continual-rl.bsky.social bsky.app/profile/cogs...
bsky.app
1201
Reposted by @cogscikid.bsky.social
Charley Wu @thecharleywu.bsky.social · 20/06/2026
Look how far we've come in just a few years! We're fast as well! It turns out, recruiting reviewers is much easier when they're not being used as free labor for massively profitable publishers Support free-to-publish and Open Access science by submitting to Open Mind today! direct.mit.edu/opmi
direct.mit.edu
Open Mind | MIT Press
06422
Reposted by @cogscikid.bsky.social
Andrew Lampinen @lampinen.bsky.social · 12/06/2026
Pleased to share that this work is now published in TMLR! openreview.net/forum?id=RuW...
openreview.net
Latent learning: episodic memory complements parametric learning by...
When do machine learning systems fail to generalize, and what mechanisms could improve their generalization? Here, we draw inspiration from cognitive science to argue that one weakness of...
1346
Reposted by @cogscikid.bsky.social
Taylor Webb @taylorwwebb.bsky.social · 11/06/2026
Very excited to share that my lab will be moving to Princeton (Neuroscience & Psychology) this fall. I'll be recruiting at all levels (more info to come soon), please share / get in touch if you're interested in joining!
1110721
Reposted by @cogscikid.bsky.social
Deon T. Benton @deontbenton.bsky.social · 08/06/2026
Newborns' preference for faces is learned rather than based on a innate face-like template academic.oup.com/chidev/artic...
academic.oup.com
What can face preference in newborns tell us about their face representation? Contrasting learned familiarity with inherent bias accounts
Abstract. Possible mechanisms behind newborn face preferences are debated, including innate templates, top-heavy bias, focus on eyes, and rapid learning ex
0144
Reposted by @cogscikid.bsky.social
Mike Frank @mcxfrank.bsky.social · 04/06/2026
@manybabies.org in Nature! www.nature.com/articles/d41...
nature.com
Can an army of babies and dogs rescue psychology from its reproducibility crisis?
Labs are teaming up to run a host of huge projects that aim to bring rigour to cognitive science.
24614
cogscikid.bsky.social @cogscikid.bsky.social · 03/06/2026
do you have any papers on this?? sounds interesting!
110
cogscikid.bsky.social @cogscikid.bsky.social · 03/06/2026
no worries! diagram is welcome if you make one! very interested in simple ways that might mitigate this, especially if its in how data is processed
010
cogscikid.bsky.social @cogscikid.bsky.social · 03/06/2026
Good idea! This would be challenging. There are 3 million+ unique topologies. Are you imagining sampling 1 topology at a time during training?
120
cogscikid.bsky.social @cogscikid.bsky.social · 03/06/2026
interesting question. do you mean if policies had an inductive bias for capturing the structure of the task? If that's what you mean, probably!
110
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
If you want to learn more, or see the rest of our analysis, please check out our paper! 📄 Paper: arxiv.org/abs/2606.00880 🌐 Project page: cogscikid.com/banyan/ This was a fun collaboration with Purab Seth, Neil Shah, @gershbrain.bsky.social, @kjha02.bsky.social, & @maxkw.bsky.social !
arxiv.org
Task diversity produces systematic transfer but inhibits continual reinforcement learning
Continual reinforcement learning aims to produce agents that learn not only to improve at their current tasks but also to adapt as task distributions change. Training an agent on many diverse tasks ca...
070
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
Strikingly, higher diversity improves backward transfer to the very first distribution! So the agent keeps getting better on old tasks even as it stops improving on new ones. This suggests it's learning the shared structure, but losing the ability to specialize.
151
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
Task diversity inhibits continual reinforcement learning! At medium diversity (16 layouts × 16 topologies), performance climbs with each new distribution. But past that point, it monotonically gets worse.
120
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
So what happens across a sequence of ten distributions? We co-vary the number of layouts with the number of topologies, and find...
130
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
We analyze backward transfer and find that varying the number of layouts or object assignments has no effect, but *varying the number of topologies has a big effect*! Again, this is true for both PPO and PQN.
130
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
Vary any single component and agents begin a new distribution at or above the same performance that they finished the previous one with! This is true for both PPO and PQN.
140
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
So we built Banyan: a procedurally-generated environment where you can systematically vary the components that define a task. A Banyan task is composed of a layout & a task tree, where each tree is defined by its topology and the objects assigned to its nodes.
250
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
Interestingly, diversity has even been key to cognitive science results. Lake & Baroni showed that the right diverse task distribution induced systematicity in neural networks! A question open since Fodor & Pylyshyn. But nobody has studied what diversity does to continual learning.
150
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
In ML, diversity is the standard route to generalization. For example, domain randomization has shown that visual diversity enables zero-shot sim-to-real and XLand showed that agents trained on diverse tasks could then solve millions of new ones zero-shot.
130
cogscikid.bsky.social @cogscikid.bsky.social · 02/06/2026
Task diversity is supposedly key to generalization in RL. But what does it do to continual RL, where agents face one new task distribution after another? We find that past a point, more diversity actually inhibits continual reinforcement learning 🧵
45114
Reposted by @cogscikid.bsky.social
DuckDuckGo @duckduckgo.com · 29/05/2026
Since Google revealed its plans for an AI search overhaul, visits to our "No AI" search page have tripled…and they’re still rising! Want to make it your default on Chrome or Firefox? Grab our No-AI extensions and banish AI-assisted answers, chat, and AI images. (1/3)
316148655238
Reposted by @cogscikid.bsky.social
Mike Frank @mcxfrank.bsky.social · 26/05/2026
If a student reads a typical psychology paper, picks an experiment, and tries to replicate it, they have roughly a coin-flip chance of success. That's the punchline of a decade of metascience, and it's the focus of Ch 3 of Experimentology. 🧵 experimentology.io
4336
Reposted by @cogscikid.bsky.social
Andrew Lampinen @lampinen.bsky.social · 26/05/2026
We've updated the preprint of our Naturalistic Computational Cognitive Science paper (arxiv.org/abs/2502.20349) — we've tried to clarify and streamline the arguments, and added some new examples: 1/5
arxiv.org
Naturalistic Computational Cognitive Science: Towards generalizable models and theories that capture the full range of natural behavior
How can cognitive science build generalizable theories that span the full scope of natural situations and behaviors? We argue that progress in Artificial Intelligence (AI) offers timely opportunities ...
13414
cogscikid.bsky.social @cogscikid.bsky.social · 07/05/2026
Yes! And our results are used to both improve AI *and* explain human behavior!
000
cogscikid.bsky.social @cogscikid.bsky.social · 07/05/2026
thank you for this advice!
110
cogscikid.bsky.social @cogscikid.bsky.social · 07/05/2026
📄 arxiv: arxiv.org/abs/2507.05561 🌐Project page: cogscikid.com/preplay 💻 Code: github.com/wcarvalho/mu... 📊 Data + preregs: · doi.org/10.17605/OSF... (grid-worlds) ·http://doi.org/10.17605/OSF.IO/B2EVM (Craftax)
arxiv.org
Preemptive Solving of Future Problems: Multitask Preplay in Humans and Machines
Humans can pursue a near-infinite variety of tasks, but typically can only pursue a small number at the same time. We hypothesize that humans leverage experience on one task to preemptively learn solu...
120
cogscikid.bsky.social @cogscikid.bsky.social · 07/05/2026
Joint work with @gershbrain.bsky.social, Honglak Lee, Sam Hall-McMaster Paper + code + data ↓
000
cogscikid.bsky.social @cogscikid.bsky.social · 07/05/2026
Scaling there will lean on two corrections we developed along the way: off-task Q(λ) and conservative all-goals learning. They're what let us match the human data — and my bet for what will unlock Habitat-class transfer with pure RL.
100