A study of animal action segmentation algorithms across supervised, unsupervised, and semi-supervised learning paradigms
Neurons, Behavior, Data analysis, and Theory · 2024
DOI: 10.51628/001c.127770 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neurons, Behavior, Data analysis, and Theory, 2024. Verfasst von Ari Blau und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics, Columbia University. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Wellcome Trust, NINDS NIH HHS.
Abstract (Originalsprache)
Action segmentation of behavioral videos is the process of labeling each frame as belonging to one or more discrete classes, and is a crucial component of many studies that investigate animal behavior. A wide range of algorithms exist to automatically parse discrete animal behavior, encompassing supervised, unsupervised, and semi-supervised learning paradigms. These algorithms – which include tree-based models, deep neural networks, and graphical models – differ widely in their structure and assumptions on the data. Using four datasets spanning multiple species – fly, mouse, and human – we systematically study how the outputs of these various algorithms align with manually annotated behaviors of interest. Along the way, we introduce a semi-supervised action segmentation model that bridges the gap between supervised deep neural networks and unsupervised graphical models. We find that fully supervised temporal convolutional networks with the addition of temporal information in the observations perform the best on our supervised metrics across all datasets.
Autor:innen und Institutionen
- Department of Statistics, Columbia University.
- Icahn School of Medicine, Mount Sinai.
- Zuckerman Institute, Columbia University.
- Sainsbury Wellcome Centre, University College London.
Förderung
- Wellcome Trust
- NINDS NIH HHS
Zitieren
APA
Blau, A., Schaffer, E. S., Mishra, N., Miska, N. J., International Brain Laboratory, Paninski, L., & Whiteway, M. R. (2024). A study of animal action segmentation algorithms across supervised, unsupervised, and semi-supervised learning paradigms. Neurons, Behavior, Data analysis, and Theory. https://doi.org/10.51628/001c.127770
Vancouver
Blau A, Schaffer ES, Mishra N, Miska NJ, International Brain Laboratory, Paninski L, et al. A study of animal action segmentation algorithms across supervised, unsupervised, and semi-supervised learning paradigms. Neurons, Behavior, Data analysis, and Theory. 2024. doi:10.51628/001c.127770
BibTeX
@article{Blau2024,
author = {Blau, Ari and Schaffer, Evan S and Mishra, Neeli and Miska, Nathaniel J and International Brain Laboratory, International Brain Laboratory and Paninski, Liam and Whiteway, Matthew R},
title = {A study of animal action segmentation algorithms across supervised, unsupervised, and semi-supervised learning paradigms},
journal = {Neurons, Behavior, Data analysis, and Theory},
year = {2024},
doi = {10.51628/001c.127770}
}
EUVIMED