Salvato in:
Dettagli Bibliografici
Autori principali: Martin, Jr., Donald, Kinney, David
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2309.10211
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910740307247104
author Martin, Jr., Donald
Kinney, David
author_facet Martin, Jr., Donald
Kinney, David
contents Deep learning is a powerful set of techniques for detecting complex patterns in data. However, when the causal structure of that process is underspecified, deep learning models can be brittle, lacking robustness to shifts in the distribution of the data-generating process. In this paper, we turn to loop polarity analysis as a tool for specifying the causal structure of a data-generating process, in order to encode a more robust understanding of the relationship between system structure and system behavior within the deep learning pipeline. We use simulated epidemic data based on an SIR model to demonstrate how measuring the polarity of the different feedback loops that compose a system can lead to more robust inferences on the part of neural networks, improving the out-of-distribution performance of a deep learning model and infusing a system-dynamics-inspired approach into the machine learning development pipeline.
format Preprint
id arxiv_https___arxiv_org_abs_2309_10211
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Loop Polarity Analysis to Avoid Underspecification in Deep Learning
Martin, Jr., Donald
Kinney, David
Machine Learning
Human-Computer Interaction
Methodology
Deep learning is a powerful set of techniques for detecting complex patterns in data. However, when the causal structure of that process is underspecified, deep learning models can be brittle, lacking robustness to shifts in the distribution of the data-generating process. In this paper, we turn to loop polarity analysis as a tool for specifying the causal structure of a data-generating process, in order to encode a more robust understanding of the relationship between system structure and system behavior within the deep learning pipeline. We use simulated epidemic data based on an SIR model to demonstrate how measuring the polarity of the different feedback loops that compose a system can lead to more robust inferences on the part of neural networks, improving the out-of-distribution performance of a deep learning model and infusing a system-dynamics-inspired approach into the machine learning development pipeline.
title Loop Polarity Analysis to Avoid Underspecification in Deep Learning
topic Machine Learning
Human-Computer Interaction
Methodology
url https://arxiv.org/abs/2309.10211