Talmo Pereira

Hope to see you at the @neuripsconf.bsky.social @dataonbrainmind.bsky.social Workshop! I will presenting our poster “Massively Parallel Imitation Learning of Mouse Forelimb Musculoskeletal Reaching Dynamics” @talmo.bsky.social @eazim.bsky.social Dec 7, SDCC 🦾🧠
arxiv.org/abs/2511.21848

🚨Our work was accepted to the @neuripsconf.bsky.social : Data on the Brain & Mind workshop!🧠🦾"Massively Parallel Imitation Learning of Mouse Forelimb Musculoskeletal Reaching Dynamics" @talmo.bsky.social @eazim.bsky.social
An imitation learning framework for modeling mouse forelimb control. 1/3

Talmo Lab is at @sfn.org! Come check out our latest work!

#neuroskyence #neurosky #SfN2025 #SfN25

Getting the most of out a scientific conference A slightly unconventional guide to your first time

Great tips from @hormiga.bsky.social for your first scientific conference — in perfect time for #SfN2025!

#neuroskyence

scienceforeveryone.science/getting-the-...

Dark blue graphic with speaker headshot reads, "October 14 | Session Spotlight | Quantifying and modeling behavior using deep learning | Talmo Pereira, Ph.D. | Salk Institute for Biological Studies | 2025 Machine Learning Short Course | October 12 - 17 | Bar Harbor, ME & Virtual."

📣 SPEAKER SPOTLIGHT: Learn about quantifying and modeling behavior using #DeepLearning from @talmo.bsky.social.

➡️ Apply to attend: www.jax.org/education-an...

#MachineLearning #Neuroskyence

Super excited to share the first cooperative foraging paradigm in freely interacting mouse pairs! Stable leader and follower roles emerge spontaneously and predict learning. Well-trained mice show stereotyped, role-specific “behavioral motifs” absent in naive animals (1/5)
doi.org/10.1101/2025...

https://www.kaggle.com/competitions/MABe-mouse-behavior-detection

And we are live!

Excited to announce the 2025 Multi-Agent Behavior Challenge on cross-lab supervised action recognition in mice 🐁🐀🖱️

Running on Kaggle until December 15th, with a $50,000 prize pool going to the top five submissions!

www.kaggle.com/competitions...

📽️Recordings from our
@cosynemeeting.bsky.social
#COSYNE2025 workshop on “Agent-Based Models in Neuroscience: Complex Planning, Embodiment, and Beyond" are now online: neuro-agent-models.github.io
🧠🤖

I recently spoke at a Congressional briefing with a military veteran from our PTSD clinical trial on the life-changing impact of the NIH Brain Initiative. Watch here👇

braininitiative.nih.gov/news-events/...

The BRAIN Initiative is now facing a 20% cut on top of the 40% last yr…

Having funded teams/contracts that required six figure+ funds a few times now, these numbers SOUND really huge in isolation to normal folks but they're not! Don't make floating numbers! I like to say things like this: "A cancer lab is basically a small business, with many jobs created by each grant"

DISCOVER Symposium - Salk Institute for Biological Studies Applications are open! Program Overview & Goals The Salk Institute for Biological Sciences Symposium for a Diverse Inclusive Scientific Community Offering a Vision for an Ecosystem Reimagined (DISCO...

This is an amazing opportunity for graduate students in #PlantScience who consider a postdoc. 3-day, all-expenses-paid program from July 23-25, 2025 in
San Diego! Check it out and spread the word!
www.salk.edu/about/our-co...

Announcing a new week-long program for young computational neuroscience/ behavior professors to talk about rigorous science, mentoring, lab management, and networking in a stunning retreat setting. Do great science as a community and have fun doing so.

Today was the conclusion of the SLEAP.ai Hackathon 2024 in @talmo.bsky.social Lab! This year we improved SLEAP deep learning software by fixing issues, answering discussions, improving documentation and adding features! Thanks to Liezl, Elizabeth,Talmo and the SLEAP dev team for making it happen!

Open science: hope is other people - Change, Technically Much like open source software, open science is a path to distributed collaboration. By sharing the data from experiments and investigations open and available, scientists can multiply impact and disc...

If we want science and software to be more open, we have to learn how to communicate and collaborate.

It was a joy to chat with @sejdevries.bsky.social from @alleninstitute.bsky.social) about what it will take to make science truly open and team-based.

www.changetechnically.fyi/2396236/epis...

Parallel development of social behavior in biological and artificial fish - Nature Communications The learning algorithms that produce social behavior are unknown. Here the authors show that artificial fish, modeled via embodied deep neural networks, learn the same social behaviors as real fish, s...

Another cool "digital twin" work from Wood’s lab

After the virtual chicks, this time:
"We embodied artificial neural networks in artificial fish and raised the artificial fish in virtual fish tanks that mimicked the rearing conditions of biological fish"

#NeuroAI

www.nature.com/articles/s41...

Cosyne 2025 Undergraduate Travel Grant Program

The Undergrad Travel Grant Program (deadline: Dec 6th) provides an opportunity for undergrads to learn more about comp systems neuro. It’s especially suited for students considering neuro graduate study!

Apply here:
docs.google.com/forms/d/e/1F...

More info:
www.cosyne.org/travel-grants

There isn't yet BUT the BRAIN Initiative BBQS just funded two key projects to fill this gap:

BARD.CC (reporter.nih.gov/project-deta... ): Data and software coordinating center
and
EMBER (reporter.nih.gov/project-deta... ): A behavior-oriented data archive

They're both by folks in the DANDI network

E11 Bio Roadmap | E11 Bio

"E11 Bio is excited to share a major step towards brain mapping at 100x lower cost, making whole-brain connectomics at human & mouse scale feasible (🧠→🔬→💻). Critical for curing brain disorders, building human-like AI systems, and even simulating human brains." -CEO Andrew Payne e11.bio/news/roadmap

Great example of “last mile” tooling needed to bridge AI to scientific workflows.

It’s hard to get these built in academia because they require a traditional software eng skillset (eg. building UIs) that scientists rarely have.

🧪 E11 Bio is excited to share a major step towards brain mapping at 100x lower cost, making whole-brain connectomics at human & mouse scale feasible (🧠→🔬→💻). Critical for curing brain disorders, building human-like AI systems, and even simulating human brains.

Read more: e11.bio/news/roadmap

In case you haven't seen it yet, movement is one of the most interesting and well engineered toolkits working with pose tracking data like from sleap.ai!

Strong recommend 🤙

Sniffing helps animals identify smells and connect them to places and events, but noses can’t sense time or place.

How do brains connect odors with internal models of the world?

Our preprint suggests that the olfactory bulb participates in this connection.

www.biorxiv.org/content/10.1...

This is my first post here, and I'm happy to share some news. We've just launched megabouts.ai! 🐟📈
1/n

Just In case people need to follow people in neuroscience focused on natural behaviors and the brains of cool non-trad animals! Please join let me know if you wanna be added. go.bsky.app/AcLpM4h

Clusters of behavior generated with SLEAP tool

4/ 🛠 To dig deeper, we used machine learning (@talmo.bsky.social SLEAP 🧠🤖) to track reward competition behavior. It revealed hidden differences like CD1 mice pressing against the reward port more than C57. These subtle behaviors could shape how we understand social competition.

diagrams of Behavioral assays tested

1/✨ New paper in #eNeuro! We looked at how dominance behaviors vary across mouse strains—turns out, it’s not one-size-fits-all!
Comparing CD1 & C57 male mice, we found strain-specific dominance behaviors that could change how we study social hierarchy in the lab…🧵
Paper here: tinyurl.com/bdeuf6s9