NeurIPS Sensorium Competition workshop is happening on Friday from 1:30–4:30 in room 357. Explore with us the state of the art and future of predictive models of neural responses. Keynotes by @dyamins, Colin Conwell and @s_y_chung

In addition to the main track for predicting responses to natural videos, this year we test how well models generalize in a bonus track with out-of-domain videos.  3/7

This year’s competition includes a brand new, large-scale dataset with responses of over >38,000 neurons in mouse V1 to hundreds of videos 📽️, replacing the static images from 2022. How you handle the new temporal dimension ⌛️ will be a major part of this year’s challenge! 2/7

Do you want to understand how the brain 🧠 dynamically processes visual stimuli? Join #Sensorium2023 at #NeurIPS2023, where your dynamic model competes 🏆 to predict tens of thousands of neurons in early mouse visual system! 🐭 1/7

We provide an extensive starting kit: github.com/sinzlab/sensorium. It features demo notebooks on how to load the data, fit a neural predictive model, and make a competition submission. For inspiration, we added notebooks to explain our pre-trained models and how they learn retinotopy