Saved in:
Bibliographic Details
Main Authors: Sristi, Ram Dyuthi, Lindenbaum, Ofir, Lifshitz, Shira, Lavzin, Maria, Schiller, Jackie, Mishne, Gal, Benisty, Hadas
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.14254
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909219529162752
author Sristi, Ram Dyuthi
Lindenbaum, Ofir
Lifshitz, Shira
Lavzin, Maria
Schiller, Jackie
Mishne, Gal
Benisty, Hadas
author_facet Sristi, Ram Dyuthi
Lindenbaum, Ofir
Lifshitz, Shira
Lavzin, Maria
Schiller, Jackie
Mishne, Gal
Benisty, Hadas
contents Feature selection is a crucial tool in machine learning and is widely applied across various scientific disciplines. Traditional supervised methods generally identify a universal set of informative features for the entire population. However, feature relevance often varies with context, while the context itself may not directly affect the outcome variable. Here, we propose a novel architecture for contextual feature selection where the subset of selected features is conditioned on the value of context variables. Our new approach, Conditional Stochastic Gates (c-STG), models the importance of features using conditional Bernoulli variables whose parameters are predicted based on contextual variables. We introduce a hypernetwork that maps context variables to feature selection parameters to learn the context-dependent gates along with a prediction model. We further present a theoretical analysis of our model, indicating that it can improve performance and flexibility over population-level methods in complex feature selection settings. Finally, we conduct an extensive benchmark using simulated and real-world datasets across multiple domains demonstrating that c-STG can lead to improved feature selection capabilities while enhancing prediction accuracy and interpretability.
format Preprint
id arxiv_https___arxiv_org_abs_2312_14254
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Contextual Feature Selection with Conditional Stochastic Gates
Sristi, Ram Dyuthi
Lindenbaum, Ofir
Lifshitz, Shira
Lavzin, Maria
Schiller, Jackie
Mishne, Gal
Benisty, Hadas
Machine Learning
Neural and Evolutionary Computing
Feature selection is a crucial tool in machine learning and is widely applied across various scientific disciplines. Traditional supervised methods generally identify a universal set of informative features for the entire population. However, feature relevance often varies with context, while the context itself may not directly affect the outcome variable. Here, we propose a novel architecture for contextual feature selection where the subset of selected features is conditioned on the value of context variables. Our new approach, Conditional Stochastic Gates (c-STG), models the importance of features using conditional Bernoulli variables whose parameters are predicted based on contextual variables. We introduce a hypernetwork that maps context variables to feature selection parameters to learn the context-dependent gates along with a prediction model. We further present a theoretical analysis of our model, indicating that it can improve performance and flexibility over population-level methods in complex feature selection settings. Finally, we conduct an extensive benchmark using simulated and real-world datasets across multiple domains demonstrating that c-STG can lead to improved feature selection capabilities while enhancing prediction accuracy and interpretability.
title Contextual Feature Selection with Conditional Stochastic Gates
topic Machine Learning
Neural and Evolutionary Computing
url https://arxiv.org/abs/2312.14254