Alireza Modirshanechi

We are looking for PhD students via the @ellis.eu and @munichcenterml.bsky.social PhD programs!

Exploration Patterns in Spontaneous Self-Touch Actions in Infancy Infants frequently touch their own bodies from the earliest months of life, raising questions about whether these self-directed actions reflect active body exploration. We hypothesize that infants’ se...

We have a new conference paper out just now! How do infants explore their own body? We show that in the first weeks of life, infants' self-touch actions allow them to gain information about their own body. ieeexplore.ieee.org/document/112...

How do we weigh curiosity, effort, and reward? The labs of @epfl-brainmind.bsky.social professors Wulfram Gerstner & Michael Herzog find humans overexplore when seeking monetary reward, drawn to novel options even when more exploration offers no advantage. Explore the paper 👉 doi.org/10.1073/pnas...

The Mere Presence of a Cell Phone May be Distracting: Implications for Attention and Task Performance: Social Psychology: Vol 45, No 6

Sure! econtent.hogrefe.com/doi/10.1027/...

I originally found it from this review, maybe that’s also interesting: www.frontiersin.org/journals/psy...

Thanks a lot! :)

This one is so interesting! Would you mind sharing the reference? :)

Decided to challenge myself to share more often online without filtering and overthinking, so starting a 365-day challenge where I’ll post a neuro fact once a day.

#ShyButSharing365 (Feel free to join, doesn’t have to be neuro-related)

Fully-funded 4-year #PhD in Cultural Evolution! Join my @erc.europa.eu project exploring how compression & compositionality drive cultural innovation: hmc-lab.com/ERCPhDCultur...
Apply by Nov 12!
Maybe of interest to folks from #COSMOS2025 or @eslr.bsky.social? Please feel free to share! 🙏

Very thought-provoking post by @prakhargodara.bsky.social. Is confirmation bias/positivity bias a statistical "ghost" of model specification? Specifically not including temporally decaying learning rates? The evidence suggests this is not the case and here is why (1/n)

🎓 Alumni Spotlight: @mmondelli.bsky.social
From EPFL to @istaresearch.bsky.social , Marco’s journey in Algorithms & Theoretical Computer Science shows the impact of an EDIC PhD. Today, he’s advancing research in algorithms, coding theory, and data science.

Over the past year, my lab has been working on fleshing out theory + applications of the Platonic Representation Hypothesis.

Today I want to share two new works on this topic:

Eliciting higher alignment: arxiv.org/abs/2510.02425
Unpaired learning of unified reps: arxiv.org/abs/2510.08492

1/9

Science is a question-answering activity. Whenever someone tries to pose a scientific question about art, it always turns out to really be a question about something else (perception, emotion, etc). This is why many attempts to bring art and science together fall flat.

If you're interested in doing a postdoc at @icepfl.bsky.social , there's still time to apply for the @epfl-ai-center.bsky.social postdoctoral fellowships.

Apart from this, I'm also recruiting postdocs in developing novel training algorithms for reasoning models and agentic AI.

Perfect. Thanks! :)

Last week, we had three amazing days full of discussions and wondering what curiosity actually means. Thank you to @modirshanechi.bsky.social, @mjgruber.bsky.social, and Lisa Beinborn for being there with us, it was a pleasure. Special thanks to Azzurra Ruggeri for her great talk!

I see, I see. Then I guess that can also be settled by doing a large-scale model comparison, across datasets, to test symmetric dynamic learning rate against asymmetric fixed learning rates; has anything similar been done before? :-?

I think that’s exactly where I interpret it differently: the Bayes solution has a *symmetric* but *dynamic* learning rate. If we (incorrectly) assume *fixed* learning rates when modeling its behavior, we end up inferring *asymmetry* that isn’t really there.

This is really cool! But I read the paper a bit differently. I see it more as arguing that evidence for asymmetric updates in human RL might be a result of model mis-specification (i.e., constant learning rates), rather than claiming asymmetric updates emerge from optimality. Am I missing something?

A glimpse at what #NeuroAI brain models might enable: a topographic vision model predicts stimulation patterns that steer complex object recognition behavior in primates. This could be a key 'software' component for visual prosthetic hardware 🧠🤖🧪

🧠 New preprint: we show that model-guided microstimulation can steer monkey visual behavior.

Paper: arxiv.org/abs/2510.03684

🧵

After a few days of extensive discussions about curiosity at @rtg2906-curiosity.bsky.social's retreat, I realized—perhaps for the first time—that the word curiosity in Persian (کنجکاوی) literally means "exploring corners"! It's such a vivid way to express curiosity! so amusing to think about it!

Golden eagle on the nest in Finland (by O. Karlin)

🦅PhD position 🦅 in my new group at @fbm-unil.bsky.social in Switzerland, studying how the social and resource landscapes shape the learning process for soaring flight. Deadline: Oct 30. Pls repost! career5.successfactors.eu/career?caree...

I had so much fun organising #CISE2025 conference on Curiosity, Information Seeking & Exploration with Romy Frömer, @hayleydorfman.bsky.social, Ohad Dan, Matt Nassar, Tali Sharot & Jacqueline Gottlieb at beautiful Brown University then hearing about all the amazing research from our speakers…

Leveling up fun: learning progress, expectations, and success influence enjoyment in video games Scientific Reports - Leveling up fun: learning progress, expectations, and success influence enjoyment in video games

What influences whether people have fun with a task?

Our paper “Leveling up fun: learning progress, expectations and success influence enjoyment in video games” with @thecharleywu.bsky.social and @ericschulz.bsky.social now in Scientific Reports!

rdcu.be/eI069

Paper summary below 1/4

The neural control & computation lab is recruiting!

If you're interested in using large-scale neural population recordings to study how the brain learns to produce complex and flexible behaviours, please get in touch.

www.ncclab.ca

Thank you! :)

Tour de force talk decomposing aspects of novelty, surprise, info gain in multi-state tasks by @modirshanechi.bsky.social

Novelty-seeking (visit new states) is common...even when its suboptimal
#CISE2025

Cartoon showing a stick figure looking toward an end goal w/ a bumpy path to it.

Wise cartoon posted outside a @princetonneuro.bsky.social PhD student office. Let’s all remember to enjoy the journey today (and all days).

Also happy to announce that our Automated scientific minimization of regret paper got accepted to the AI4Science workshop at #NeurIPS - arxiv.org/abs/2505.17661 with @marcelbinz.bsky.social, @akjagadish.bsky.social & @ericschulz.bsky.social

We are organizing a workshop on Metacognition in Generative AI at @euripsconf.bsky.social in Copenhagen later this year.
Submission deadline for short papers is on October 17th.