Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster F in Poster Session F: Thursday, August 6, 1:45 – 3:30 pm, Kimmel Center, Shorin & Rosenthal Rooms

Discovering interpretable visual representations at scale

Habon Issa¹, Sunny Liu¹, David Klindt¹; ¹Cold Spring Harbor Laboratory

Presenter: Habon Issa

Despite large-scale recordings, neuroscience has identified representations for only a small fraction of visual features. We address this by extracting high-dimensional population codes from biological and artificial vision networks, then quantifying the interpretability and diversity of neuron vs. population representations at scale using automated evaluations that mirror human judgments. Population codes and automated analyses together reveal more interpretable, diverse representations. Our framework meaningfully closes the gap between the scale of neural recordings and the depth of insight they yield, motivating a broader shift away from dimensionality reduction in studies of visual perception.

Topic Area: Methods, Tools, Theory & Neural Coding