What Do You See in Common? Learning Hierarchical Prototypes over Tree-of-Life to Discover Evolutionary Traits

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Manogaran, Harish Babu, Maruf, M., Daw, Arka, Mehrab, Kazi Sajeed, Charpentier, Caleb Patrick, Uyeda, Josef C., Dahdul, Wasila, Thompson, Matthew J, Campolongo, Elizabeth G, Provost, Kaiya L, Chao, Wei-Lun, Berger-Wolf, Tanya, Mabee, Paula M., Lapp, Hilmar, Karpatne, Anuj
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913894466846720
author Manogaran, Harish Babu
Maruf, M.
Daw, Arka
Mehrab, Kazi Sajeed
Charpentier, Caleb Patrick
Uyeda, Josef C.
Dahdul, Wasila
Thompson, Matthew J
Campolongo, Elizabeth G
Provost, Kaiya L
Chao, Wei-Lun
Berger-Wolf, Tanya
Mabee, Paula M.
Lapp, Hilmar
Karpatne, Anuj
author_facet Manogaran, Harish Babu
Maruf, M.
Daw, Arka
Mehrab, Kazi Sajeed
Charpentier, Caleb Patrick
Uyeda, Josef C.
Dahdul, Wasila
Thompson, Matthew J
Campolongo, Elizabeth G
Provost, Kaiya L
Chao, Wei-Lun
Berger-Wolf, Tanya
Mabee, Paula M.
Lapp, Hilmar
Karpatne, Anuj
contents A grand challenge in biology is to discover evolutionary traits - features of organisms common to a group of species with a shared ancestor in the tree of life (also referred to as phylogenetic tree). With the growing availability of image repositories in biology, there is a tremendous opportunity to discover evolutionary traits directly from images in the form of a hierarchy of prototypes. However, current prototype-based methods are mostly designed to operate over a flat structure of classes and face several challenges in discovering hierarchical prototypes, including the issue of learning over-specific prototypes at internal nodes. To overcome these challenges, we introduce the framework of Hierarchy aligned Commonality through Prototypical Networks (HComP-Net). The key novelties in HComP-Net include a novel over-specificity loss to avoid learning over-specific prototypes, a novel discriminative loss to ensure prototypes at an internal node are absent in the contrasting set of species with different ancestry, and a novel masking module to allow for the exclusion of over-specific prototypes at higher levels of the tree without hampering classification performance. We empirically show that HComP-Net learns prototypes that are accurate, semantically consistent, and generalizable to unseen species in comparison to baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2409_02335
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle What Do You See in Common? Learning Hierarchical Prototypes over Tree-of-Life to Discover Evolutionary Traits
Manogaran, Harish Babu
Maruf, M.
Daw, Arka
Mehrab, Kazi Sajeed
Charpentier, Caleb Patrick
Uyeda, Josef C.
Dahdul, Wasila
Thompson, Matthew J
Campolongo, Elizabeth G
Provost, Kaiya L
Chao, Wei-Lun
Berger-Wolf, Tanya
Mabee, Paula M.
Lapp, Hilmar
Karpatne, Anuj
Computer Vision and Pattern Recognition
A grand challenge in biology is to discover evolutionary traits - features of organisms common to a group of species with a shared ancestor in the tree of life (also referred to as phylogenetic tree). With the growing availability of image repositories in biology, there is a tremendous opportunity to discover evolutionary traits directly from images in the form of a hierarchy of prototypes. However, current prototype-based methods are mostly designed to operate over a flat structure of classes and face several challenges in discovering hierarchical prototypes, including the issue of learning over-specific prototypes at internal nodes. To overcome these challenges, we introduce the framework of Hierarchy aligned Commonality through Prototypical Networks (HComP-Net). The key novelties in HComP-Net include a novel over-specificity loss to avoid learning over-specific prototypes, a novel discriminative loss to ensure prototypes at an internal node are absent in the contrasting set of species with different ancestry, and a novel masking module to allow for the exclusion of over-specific prototypes at higher levels of the tree without hampering classification performance. We empirically show that HComP-Net learns prototypes that are accurate, semantically consistent, and generalizable to unseen species in comparison to baselines.
title What Do You See in Common? Learning Hierarchical Prototypes over Tree-of-Life to Discover Evolutionary Traits
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.02335