Séminaire Paul Expert (Imperial College London) : Localisation in persistent homology
May 02, 2018
from 02:00 to 03:00
|Where||ENS de Lyon, site Monod, salle 116 (MGN1)|
|Add event to calendar||
Localisation in persistent homology
Summary: Persistent homology is a recent method from computational topology used to characterize data by its "shape". It essentially amounts to creating a series of embeddings in the form of simplicial complexes and track the evolution their homology. An example from neuroscience will be used to illustrate the importance of the localization of homology in characterizing states of altered consciousness as well as providing a way to interpret the results of persistent homology in the language of networks.
Finally, I will present some of our recent work using the spectral properties of the k-Laplacians of a simplicial complex and their relation to persistent homology.