Volltext-Downloads (blau) und Frontdoor-Views (grau)

How inverse conditional flows can serve as a substitute for distributional regression

  • Neural network representations of simple models, such as linear regression, are being studied increasingly to better understand the underlying principles of deep learning algorithms. However, neural representations of distributional regression models, such as the Cox model, have received little attention so far.We close this gap by proposing a framework for distributional regression using inverse flow transformations (DRIFT), which includes neural representations of the aforementioned models. We empirically demonstrate that the neural representations of models in DRIFT can serve as a substitute for their classical statistical counterparts in several applications involving continuous, ordered, timeseries, and survival outcomes. We confirm that models in DRIFT empirically match the performance of several statistical methods in terms of estimation of partial effects, prediction, and aleatoric uncertainty quantification. DRIFT covers both interpretable statistical models and flexible neural networks opening up new avenues in both statistical modeling and deep learning.

Export metadata

Additional Services

Search Google Scholar

Statistics

frontdoor_oas
Metadaten
Author:Lucas Kook, Chris Kolb, Philipp Schiele, Daniel Dold, Marcel ArpogausORCiD, Cornelius Fritz, Philipp F. Baumann, Philipp Kopper, Tobias Pielok, Emilio Dorigatti, David Rügamer
URL:https://dl.acm.org/doi/10.5555/3702676.3702771
DOI:https://doi.org/10.48550/arXiv.2405.05429
ISSN:2640-3498
Parent Title (English):Proceedings of Machine Learning Research (PMLR)
Volume:Vol. 244
Publisher:JMLR
Place of publication:Cambridge, MA
Document Type:Conference Proceeding
Language:English
Year of Publication:2024
Contributing Corporation / Conference:Conference on Uncertainty in Artificial Intelligence (40th : 2024 : Barcelona, Spain)
Release Date:2025/01/28
First Page:2029
Last Page:2046
Article Number:95
Institutes:Fakultät Elektrotechnik und Informationstechnik
Open Access?:Nein
Relevance:Konferenzbeitrag: h5-Index > 30
Licence (German):License LogoUrheberrechtlich geschützt