TOPOLOGICAL LEARNING FOR BRAIN NETWORKS

• Published in: The annals of applied statistics Vol. 17; no. 1; p. 403
• Main Authors: Songdechakraiwut, Tananun, Chung, Moo K
• Format: Journal Article
• Language: English
• Published: United States 01.03.2023
• Subjects: topological learning; Wasserstein distance; twin brain imaging study; Topological data analysis; birth-death decomposition; persistent homology

This paper proposes a novel topological learning framework that integrates networks of different sizes and topology through persistent homology. Such challenging task is made possible through the introduction of a computationally efficient topological loss. The use of the proposed loss bypasses the...