Using Topology to Study the Geometry of Neural Correlations
Explore the application of topology in studying neural correlations through this insightful lecture. Delve into the concept of Betti curves of symmetric matrices, understanding their role as matrix invariants dependent on the relative ordering of matrix entries. Learn how persistent homology is used to compute these invariants and how they can reveal underlying structures in biological data that may be obscured by monotone nonlinearities. Examine previous applications of Betti curves in hippocampal and olfactory data studies. Discover new theorems characterizing Betti curves of rank 1 symmetric matrices and observe how these Betti curve signatures manifest in natural data obtained from calcium imaging of neural activity in zebrafish. Gain valuable insights into the intersection of topology, neuroscience, and data analysis in this comprehensive exploration of neural correlation geometry.