Data Science Influencer: “Alright everyone, it’s pride 🌈🌈month✨✨✨, so you know what that means: ranking the machine learning algorithms by how gay they are”

I’m just waiting for the next W.H. announcement to be “It's got what plants crave. It's got electrolytes”

#neuroskyence GLM/GAM are used for neural tuning estimation. I made updates to an easy to use python lib. Supports splines, and regs. and likelihood combos not readily available (ARD , Bayesian Group Lasso, etc), (poisson, neg binomial, gauss, zeroinf). Jax/NumPyro for high-D bayes. Linked

I could not have done this without the sweat and immense effort driven by a whole team of people. @assiachericoni.bsky.social in particular.

Excited to share our latest preprint with @camillopadoasch.bsky.social and Xiao-Jing Wang! We present a biologically plausible framework showing how neural circuits compute & compare value to drive flexible economic decision making. www.biorxiv.org/content/10.1...

If you identify as a Jewish member of the academic community, and are appalled by the actions being taken in the name of defending our community, please consider signing this letter (which can be done anonymously) and also sharing it with others: forms.gle/prnRbq69a6YN...

the authoritarianism is bad, but the loser energy is absolutely intolerable

New Lab Preprint! "Independent Continuous Tracking of Multiple Agents in the Human Hippocampus" led by my graduate student Assia Chericoni. www.biorxiv.org/content/10.1...

👏

This is really cool!

What’s the minimal amount of time I have to write before I play Elden? Asking for a friend.

Do you want to fit nonlinear tuning curves, using Poisson (GAM/GLM) models, but want flexible basis functions, Bayesian machinery, and smoothing and regularization of penalized splines. I wanted a bayesian approach in a modern framework (Pyro). Without guarantee github.com/JustFineNeur...

I would like to highlight a somewhat tangential finding here: We found clear evidence for clusters in the encoding geometry of natural images and a decision making task (in mice and monkeys). I think this speaks in favor of modular brains in both species.

🙌 this has been a tough lesson for me, as I transitioned from cog sci to sys neuro. I feel like I’m shoehorning in analysis merely for attentional capture.

My first @cosynemeeting.bsky.social meeting abstract for me submitted and accepted, nice.

most people aren’t defining their subspaces. Take the issue of PCA. If experiment cond A pushes a latent to have higher variance than condition B, comparing the two artificially thru alignment artificially inflates “orthogonaliy”. Better to define geometry based on variables and decompose.

Can the motor system use sensory expectations to prepare for unexpected events? Excited to share my latest work with @andpru.bsky.social – where we establish that sensory expectations shape neural population dynamics in motor circuits! 🧵 and paper below 1/

Fresh out: a short review article where I give an overview of different ways that across-trial temporal structure is helpful in understanding cognitive models of decision-making osf.io/preprints/ps...

Has anybody made a variant of a demixed dim red that is DS based, with high interpretability? basically pitching demixed LDS with Nonlinear approximated inputs? #neuroskyence

A lot of musing about how manifolds replace tuning. But I’d like to argue this is slight misdirection. You don’t get manifolds absent tuning/connectictivity/neuromodulation. Manifolds not in lieu of tuning, but they are independent in so far multiple tunings can render isomorphic manifolds.

I’m happy someone said this. I saw a few packs that had people on it, that don’t even do that research (but I do). But I also suck at self-promoting. Maybe gotta find the bravery to ask for a seat at the proverbial table.

I am a bit afraid the whole Starter Pack business leaves behind trainees — make sure to follow / repost them if their work interests you so their voices are heard #PsychSciSky #neuroskyence #compneurosky

Random opinion: stochastic optimal control theory is underrated in NeuroAI and theoretical neuro. It's one of the most general and useful frameworks out there. Lots of beautiful math and solid proofs. More people should apply it to neuroscience... 🧠🤖

Who made numpyro? I want to send them a gift basket.

Come see our nanosymposium on value-based decision making this afternoon, including my talk on the intrinsic value of information 📚🔎. Can't wait to hear from this stellar group! #SFN24

Will be bopping around @sfn this week. If anyone wants to chat, 🍻 or both, ping me!! Also, giving a talk on Monday.

A simple question spawned into surprising results: how does our allocation of attentional effort fluctuate during value based choice.

New from the lab! “Semi-orthogonal subspaces for value mediate a binding and generalization trade-off” led by @wjj.bsky.social and @justfineneuro.bsky.social! www.nature.com/articles/s41...

I fully support the last sentence of this abstract from @profdata.bsky.social : elifesciences.org/reviewed-pre... "...the complexity of the brain should be respected and intuitive notions of cell type, which can be misleading and arise in any complex network, should be relegated to history." 🧠📈 🧪

This looks cool

I particularly feel this way about “evidence accumulation” signals atm. I keep seeing it, but can’t convince myself it’s what I think. 🤷🏻‍♂️

The only BCI I want is the one that notates my good research ideas, which only come to me when I’m doing anything but working.

Many researchers in cognitive neuroscience and deep learning seem to assume that inference in the brain is hierarchical.  Our new Nature Reviews Neuroscience paper questions this assumption and proposes the shallow brain hypothesis. #neuroskyence #brain www.nature.com/articles/s41...

Can someone with expertise in navigation neuro (place, grid etc) help me out. What’s the difference between goal distance coding and reward proximity? Is it just you could be close to a rewarding object, but it isn’t necessarily your goal? An exemplar difference/paper would be great.

Can learning occur offline w asyncrhonous inputs? Yes, in systems alignment entire embeddings (e.g., visual and linguistic) are mapped in an unsupervised manner. Systems alignment can support children's word learning. IMO, this is an important form of learning neglected in psych, neuro, and ML.

How does the brain represent multiple different things at once in a single population of neurons? @justfineneuro.bsky.social, @benhayden.bsky.social, B Ebitz, M Yoo, and I show that it uses semi-orthogonal subspaces for each item. Preprint here: arxiv.org/abs/2309.07766 Clouds below! (1/n)

The problem with effort discounting studies is the reward. It’s not “what would you do for [insert random fruit]”. “It’s what would you do for a Klondike bar?!” Do better psychologists.

My turn: psychologists who study motor control are the bastards of psychology and dogs of biomedical engineering.

Question for my whole 10 followers: would you consider “change of mind” in decision making a form of ‘(cognitive) control’. If so/not, why? I’m not sure I yet see it as the same as automatic + controlled style thinking. Cheers!

This is the right advice. I worked too many weekends in grad school. In hindsight, it was a net loss in mental sanity and life enjoyment. Also, anime recs. are welcomed.