Lastly, if you're excited about developing interpretable machine learning tools for neural data analysis, we are recruiting postdocs - please send me an email if interested 4/4

A huge thanks to the large team involved in this project, led by twitterless Andrew Zimnik @MarkChurchland @sherryxinyuean @lndriscoll @VladislavSusoy (and several more) 3/4

Our code and example notebooks are available, and we'd love feedback if there are ways to make it more user-friendly: https://t.co/DybNa4UG4V 2/4

A brief return to twitter to share our new preprint. If you want to uncover interpretable low-dimensional factors in your neural population activity in an unsupervised way - check it out! https://t.co/7VCm0DkkSw 1/4

“I think we as scientists embody something fundamentally, beautifully human … I still, idealistically and perhaps naively, believe in beauty because we, or some of us at least, experience it and are moved by it.” — Sliman Bensmaia, PhD, 1973-2023

And a huge thanks to my many official and unofficial mentors over the years: @KordingLab, Mark, @PresNCM, @SaraASolla, Liam, John, @MarkChurchland, @scott_linderman, Alejandro, Eric, and more! 3/3

The lab will be developing interpretable machine learning tools to better understand neural activity data and collaborating heavily with the amazing local neuroscience community! 2/3

I'm very excited to announce that I will starting my lab at Northwestern Neurology @NUFeinbergMed on September 1! I will be recruiting postdocs and students, so please reach out if interested!   1/3

We urgently need mentors (post-docs or faculty) to guide ~2,000 students from around the world with their projects. Time commitment is 3 hours across 3 weeks. Familiarity with comp/systems neuro a must, but python proficiency not required. Sign up here: https://t.co/RsoCeyxLsh

Excited to announce that this is finally up on bioRxiv. Congratulations @najabam @joshuaiglase @EricMTrautmann @AmematsroElom , Sean Perkins, @shadlen Larry Abbott & John Cunningham. We hope this will change how you think about both muscles and cortex.

So great that anyone can watch the #cosyne21 talks here https://t.co/QiOqjcxICJ (good because I missed a few that I want to check out)

Coming in March: 📈 Benchmarks for latent variable models of neural population activity 📈 https://t.co/bL5kMHgeLY We're kicking this off with a modeling competition 🏆💰 Join the mailing list to stay informed! Thanks @CosyneMeeting #cosyne21 for helping us spread the word!

This is fantastic, hugely recommended for any neuroscientists getting into #python and neural #datascience https://t.co/5HzpIco7IZ

Excited (and terrified) to start releasing the material for the Math Tools for Neuroscience course I'm teaching this fall at Harvard! Khan academy style videos and tutorials/exercises in Google colab (python)

And check out our colab notebook to run our example simulation from the paper, and play around with the code yourself! https://t.co/ghNGTmVMAr   5/5

Code is now a part of @scott_linderman 's state space modeling package: https://t.co/wDamLTRXaX In particular, check out the multi-population notebook. 4/5

We demonstrate the method in simulations and in datasets from NHP motor cortex and C. elegans. 3/5

We extend recurrent switching dynamical systems models for multi-population recordings, so that in the model, populations interact via their dynamics in a lower-dimensional space. 2/5

We're excited to share our upcoming neurips paper, where we develop state space models for multi-population recordings! w/ @scott_linderman, Matt, John, and Liam https://t.co/wyi6BO47HC 1/5

Interested in how modern neural networks can be built, analyzed, and customized to study neuroscience questions? In this Primer, Xiao-Jing Wang and I provide a (hopefully) pedagogical introduction to these ideas, with equations and code https://t.co/7dLeG3LR3I