Saved in:
Bibliographic Details
Main Authors: Oono, Kenta, Charoenphakdee, Nontawat, Bito, Kotatsu, Gao, Zhengyan, Igata, Hideyoshi, Yoshikawa, Masashi, Ota, Yoshiaki, Okui, Hiroki, Akita, Kei, Yamaguchi, Shoichiro, Sugawara, Yohei, Maeda, Shin-ichi, Miyoshi, Kunihiko, Saito, Yuki, Tsuda, Koki, Maruyama, Hiroshi, Hayashi, Kohei
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.10656
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912715609473024
author Oono, Kenta
Charoenphakdee, Nontawat
Bito, Kotatsu
Gao, Zhengyan
Igata, Hideyoshi
Yoshikawa, Masashi
Ota, Yoshiaki
Okui, Hiroki
Akita, Kei
Yamaguchi, Shoichiro
Sugawara, Yohei
Maeda, Shin-ichi
Miyoshi, Kunihiko
Saito, Yuki
Tsuda, Koki
Maruyama, Hiroshi
Hayashi, Kohei
author_facet Oono, Kenta
Charoenphakdee, Nontawat
Bito, Kotatsu
Gao, Zhengyan
Igata, Hideyoshi
Yoshikawa, Masashi
Ota, Yoshiaki
Okui, Hiroki
Akita, Kei
Yamaguchi, Shoichiro
Sugawara, Yohei
Maeda, Shin-ichi
Miyoshi, Kunihiko
Saito, Yuki
Tsuda, Koki
Maruyama, Hiroshi
Hayashi, Kohei
contents Virtual Human Generative Model (VHGM) is a generative model that approximates the joint probability over more than 2000 human healthcare-related attributes. This paper presents the core algorithm, VHGM-MAE, a masked autoencoder (MAE) tailored for handling high-dimensional, sparse healthcare data. VHGM-MAE tackles four key technical challenges: (1) heterogeneity of healthcare data types, (2) probability distribution modeling, (3) systematic missingness in the training dataset arising from multiple data sources, and (4) the high-dimensional, small-$n$-large-$p$ problem. To address these challenges, VHGM-MAE employs a likelihood-based approach to model distributions with heterogeneous types, a transformer-based MAE to capture complex dependencies among observed and missing attributes, and a novel training scheme that effectively leverages available samples with diverse missingness patterns to mitigate the small-n-large-p problem. Experimental results demonstrate that VHGM-MAE outperforms existing methods in both missing value imputation and synthetic data generation.
format Preprint
id arxiv_https___arxiv_org_abs_2306_10656
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Virtual Human Generative Model: Masked Modeling Approach for Learning Human Characteristics
Oono, Kenta
Charoenphakdee, Nontawat
Bito, Kotatsu
Gao, Zhengyan
Igata, Hideyoshi
Yoshikawa, Masashi
Ota, Yoshiaki
Okui, Hiroki
Akita, Kei
Yamaguchi, Shoichiro
Sugawara, Yohei
Maeda, Shin-ichi
Miyoshi, Kunihiko
Saito, Yuki
Tsuda, Koki
Maruyama, Hiroshi
Hayashi, Kohei
Machine Learning
Artificial Intelligence
Virtual Human Generative Model (VHGM) is a generative model that approximates the joint probability over more than 2000 human healthcare-related attributes. This paper presents the core algorithm, VHGM-MAE, a masked autoencoder (MAE) tailored for handling high-dimensional, sparse healthcare data. VHGM-MAE tackles four key technical challenges: (1) heterogeneity of healthcare data types, (2) probability distribution modeling, (3) systematic missingness in the training dataset arising from multiple data sources, and (4) the high-dimensional, small-$n$-large-$p$ problem. To address these challenges, VHGM-MAE employs a likelihood-based approach to model distributions with heterogeneous types, a transformer-based MAE to capture complex dependencies among observed and missing attributes, and a novel training scheme that effectively leverages available samples with diverse missingness patterns to mitigate the small-n-large-p problem. Experimental results demonstrate that VHGM-MAE outperforms existing methods in both missing value imputation and synthetic data generation.
title Virtual Human Generative Model: Masked Modeling Approach for Learning Human Characteristics
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2306.10656