Salvato in:
Dettagli Bibliografici
Autori principali: Lee, Hyeonjun, Shin, Hyungseob, Nam, Gunhee, Lee, Hyeonsoo
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.26014
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914123902615552
author Lee, Hyeonjun
Shin, Hyungseob
Nam, Gunhee
Lee, Hyeonsoo
author_facet Lee, Hyeonjun
Shin, Hyungseob
Nam, Gunhee
Lee, Hyeonsoo
contents Survival analysis is a task to model the time until an event of interest occurs, widely used in clinical and biomedical research. A key challenge is to model patient heterogeneity while also adapting risk predictions to both individual characteristics and temporal dynamics. We propose a dual mixture-of-experts (MoE) framework for discrete-time survival analysis. Our approach combines a feature-encoder MoE for subgroup-aware representation learning with a hazard MoE that leverages patient features and time embeddings to capture temporal dynamics. This dual-MoE design flexibly integrates with existing deep learning based survival pipelines. On METABRIC and GBSG breast cancer datasets, our method consistently improves performance, boosting the time-dependent C-index up to 0.04 on the test sets, and yields further gains when incorporated into the Consurv framework.
format Preprint
id arxiv_https___arxiv_org_abs_2510_26014
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dual Mixture-of-Experts Framework for Discrete-Time Survival Analysis
Lee, Hyeonjun
Shin, Hyungseob
Nam, Gunhee
Lee, Hyeonsoo
Machine Learning
Artificial Intelligence
Survival analysis is a task to model the time until an event of interest occurs, widely used in clinical and biomedical research. A key challenge is to model patient heterogeneity while also adapting risk predictions to both individual characteristics and temporal dynamics. We propose a dual mixture-of-experts (MoE) framework for discrete-time survival analysis. Our approach combines a feature-encoder MoE for subgroup-aware representation learning with a hazard MoE that leverages patient features and time embeddings to capture temporal dynamics. This dual-MoE design flexibly integrates with existing deep learning based survival pipelines. On METABRIC and GBSG breast cancer datasets, our method consistently improves performance, boosting the time-dependent C-index up to 0.04 on the test sets, and yields further gains when incorporated into the Consurv framework.
title Dual Mixture-of-Experts Framework for Discrete-Time Survival Analysis
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.26014