Stephane Deny

New preprint! "A Deep Learning Model of Mental Rotation Informed by Interactive VR Experiments"

🔗 arxiv.org/abs/2512.13517

by Raymond Khazoum, Daniela Fernandes, Aleksandr Krylov, Qin Li and myself

@jaakkolehtinen.bsky.social

@csaalto.bsky.social

First blog post of the series explaining our motivation here: bsky.app/profile/stph...

🔥 Technical blog post (with maths and code) on how to render 4D objects, by Raihan Gafur 🔥

🔗 raihanthecooldude.com/spaceland-ep...

Nonlinear spatial integration allows the retina to detect the sign of defocus in natural scenes The retina can easily detect whether the eye is too small or too big thanks to the imperfections of the eye optics.

Happy to share my first work with a connection to myopia, a collaboration with EssilorLuxottica
www.science.org/doi/10.1126/...

How does our brain excel at complex object recognition, yet get fooled by simple illusory contours? What unifying principle governs all Gestalt laws of perceptual organization?

We may have an answer: integration of learned priors through feedback. New paper with @kenmiller.bsky.social! 🧵

@neurograce.bsky.social, @kamdh.bsky.social @csaalto.bsky.social, @aalto.fi

🔥 "The Perception and Learning of 4D Object Geometry in Humans and Machines" 🔥

Master thesis work by Raihan Gafur.

Check out his awesome blog post series!

raihanthecooldude.com/spaceland-ep...

Joint rotational invariance and adversarial training of a dual-stream Transformer yields state of the art Brain-Score for Area V4 Modern high-scoring models of vision in the brain score competition do not stem from Vision Transformers. However, in this paper, we provide evidence against the unexpected trend of Vision Transformer...

Arturo Deza had this work (arxiv.org/abs/2203.06649), but if I understand well is the kind of model you are not looking for. But Arturo may have good answers to your question.

(1/6) Thrilled to share our triple-N dataset (Non-human Primate Neural Responses to Natural Scenes)! It captures thousands of high-level visual neuron responses in macaques to natural scenes using #Neuropixels.

Training Large Neural Networks With Low-Dimensional Error Feedback Training deep neural networks typically relies on backpropagating high dimensional error signals a computationally intensive process with little evidence supporting its implementation in the brain. Ho...

(1/6) Excited to share a new preprint from our lab! Can large, deep nonlinear neural networks trained with indirect, low-dimensional error signals compete with full-fledged backpropagation? Tl;dr: Yes! arxiv.org/abs/2502.20580.

This is great stuff!

A comparison between humans and AI at recognizing objects in unusual poses

Netta Ollikka, Amro Kamal Mohamed Abbas, Andrea Perin, Markku Kilpeläinen, Stephane Deny

Action editor: Andrew Lampinen

https://openreview.net/forum?id=yzbAFf8vd5

#poses #recognition #recognizing

this would be a more apt analogy if the pilots (politicians) were piloting the planes remotely from land (politicians can be corrupt and rarely bare the consequences of their policies) and if common sense was of no relevance in politics (it is).

hmm, thanks gpt, that's very clear now what a MRAC is

new edito by N Arthaud, new banger: x.com/n_arthaud/st...

Le discours de gauche existe toujours, mais ce n'est chez les democrates americains / macronistes qu'on le trouve, et étonamment cette gauche la n'a pas quitté twitter, peut-etre justement car les classes populaires y sont mieux representee? x.com/n_arthaud/st...

My work routine summarized in two words: "2fac authentication". Thanks Microsoft 💀

New Preprint: "On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory" by Andrea Perin and myself. Retweets of this post or the thread below are highly appreciated! Follow @zazzarazzaz.bsky.social for future updates on this line of work.

😆

I agree with all said. I think the degree of control a mentor will want to exert on mentee also depends on the pressures the mentor receive to deliver (eg CV for grants / deadline for graduations / required number of publications). There is so much leeway one can have.

which journal? noteworthy information

Interesting work on symmetries and learning!

Geometry and dynamics of representations in a precisely balanced memory network related to olfactory cortex Computational modeling revealed that balanced assemblies of excitatory and inhibitory neurons shape representational manifolds in olfactory cortex-like recurrent networks, resulting in joint maps of s...

Now published in @elife.bsky.social doi.org/10.7554/eLif... a fun collaboration with
@clairemb90.bsky.social and Rainer Friedrich

I did my PhD with Ed Deci, who usually says very few words. In our meetings, it was his reserved nature that gradually allowed me to come out of my shell, be my own person. That experience teaches me one valuable lesson now that I mentor others: know when to shut up.

interesting! do you think there is a balance to find, for instance if the student seems to be taking too much risk in a project

Thanks Kameron, means a lot coming from you! Hope you enjoy the work which is certainly a product of slow science : )

New Preprint: "On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory" by Andrea Perin and myself. Retweets of this post or the thread below are highly appreciated! Follow @zazzarazzaz.bsky.social for future updates on this line of work.