FLAN: feature-wise latent additive neural models for biological applications
Briefings in Bioinformatics · 19.05.2023 · 24 (3)
DOI: 10.1093/bib/bbad056 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 06.04.2023. Verfasst von An-Phi Nguyen und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics, ETH Zürich, Rämistrasse 101, 8092 Zürich, Switzerland. Hauptthemen laut MEDLINE: Neural Networks, Computer, Machine Learning. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Horizon 2020.
Abstract (Originalsprache)
Motivation Interpretability has become a necessary feature for machine learning models deployed in critical scenarios, e.g. legal system, healthcare. In these situations, algorithmic decisions may have (potentially negative) long-lasting effects on the end-user affected by the decision. While deep learning models achieve impressive results, they often function as a black-box. Inspired by linear models, we propose a novel class of structurally constrained deep neural networks, which we call FLAN (Feature-wise Latent Additive Networks). Crucially, FLANs process each input feature separately, computing for each of them a representation in a common latent space. These feature-wise latent representations are then simply summed, and the aggregated representation is used for the prediction. These feature-wise representations allow a user to estimate the effect of each individual feature independently from the others, similarly to the way linear models are interpreted. Results We demonstrate FLAN on a series of benchmark datasets in different biological domains. Our experiments show that FLAN achieves good performances even in complex datasets (e.g. TCR-epitope binding prediction), despite the structural constraint we imposed. On the other hand, this constraint enables us to interpret FLAN by deciphering its decision process, as well as obtaining biological insights (e.g. by identifying the marker genes of different cell populations). In supplementary experiments, we show similar performances also on non-biological datasets. Code and data availability Code and example data are available at https://github.com/phineasng/flan_bio.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Mathematics, ETH Zürich, Rämistrasse 101, 8092 Zürich, Switzerland.
- IBM Research Europe, Säumerstrasse 4, 8803 Rüschlikon, Zürich, Switzerland.
Förderung
- Horizon 2020
Zitieren
APA
Nguyen, A. P., Vasilaki, S., & Martínez, M. R. (2023). FLAN: feature-wise latent additive neural models for biological applications. Briefings in Bioinformatics, 24(3). https://doi.org/10.1093/bib/bbad056
Vancouver
Nguyen AP, Vasilaki S, Martínez MR. FLAN: feature-wise latent additive neural models for biological applications. Briefings in Bioinformatics. 2023;24(3). doi:10.1093/bib/bbad056
BibTeX
@article{Nguyen2023,
author = {Nguyen, An-Phi and Vasilaki, Stefania and Martínez, María Rodríguez},
title = {FLAN: feature-wise latent additive neural models for biological applications},
journal = {Briefings in Bioinformatics},
year = {2023},
volume = {24},
number = {3},
doi = {10.1093/bib/bbad056}
}
EUVIMED