Thematic Einstein Semester on

Geometric and Topological Structure of Materials

Summer Semester 2021

Speaker


Roland Kwitt   (U Salzburg)


Title


Learning from and with persistent homology


Abstract


This talk will address the role of persistent homology in machine learning, starting from the perspective of using summary representations of topological features (i.e., barcodes) as input to machine learning methods, to more recent work on actually incorporating persistent homology within learning algorithms, neural networks in particular.

Slides



Contact


tes-summer2021@math.tu-berlin.de